Planning Your Sensory Project

PLAN YOUR
SENSORY TESTING
DR NOORAKMAR AB WAHAB
aqemanur@ums.edu.my
PREPARED BY:
AHMAD RIDUAN BAHAUDDIN
a_riduan@ums.edu.my
Planning your sensory project
1. Setting the goals and objectives

• Understand the objectives of a project - key factors in determining the
test type, experimental design and statistical analysis required
• Often, a client will want to know the answer to all of these questions;
however, time and financial constraints may require that the objectives
are prioritised.
• It is imperative that the potential outcomes of different methodologies

be highlighted in advance - aware of any limitations
2. Product type
• When selecting an appropriate methodology to satisfy the desired

objectives, it is important to consider the product type as this may have
a serious impact on test design
• When the test objective includes some aspects of performance,

products may need to be tested in the context of their use
3. Budget
• Financial constraints have to be considered in any test design.

• In some instances, the cost associated with the ‘ideal’ design exceeds the budget
and appropriate compromises such as reducing the number of products, assessors
or replicates are necessary.
• It is important to understand the consequence of every compromise with regard to
the quality of the data and the conclusions that can be drawn.
• Reducing the number of assessors from the maximum to the minimum
recommended would be acceptable
• Reducing the number of samples can be an effective way of cutting costs; however,
removing replication from the design of certain sensory methodologies, e.g.
profiling, can be very dangerous
4. Timings
• When designing a sensory test, a deadline may affect the decision over
which methodology to use → know in advance if any deadlines exist.
• In studies that require several test elements, e.g. consumer tests,

descriptive analysis, instrumental analysis, shelf life testing, the co-
ordination of these elements is crucial for samples whose sensory
properties change with time.
5. Setting action standards
• Action standards are the criteria that must be met to take a course of
action based on test results.
• They should be set in advance of the test being carried out.
• Factors to consider include size of the opportunity; business risk and
stage of testing, which will determine how strict the criteria are, as well
as product type, new or existing product category and communication
• Action standards may include number and type of consumer, statistical
criteria, and elements of the test design
6. Panellist Recruitment and Selection
• Make sure potential panellists know at need to do during the study (e.g.
commitment and product range)
• Participation – contractual relationship
➢Voluntarily participation and sign informed consent
• Screening test either by sensory test or questionnaire could be needed for certain
product evaluation
• Type of screening tests should related to the objectives/ actual evaluation
• Taste acuity assessment - basic taste, ranking test, difference test is common is
screening test
• Panellist training and performance test may needed in certain sensory test
• Before describing recruiting procedures, some guidelines on working with people as test
subjects are appropriate:-
i. Most individuals do not know what their ability is to smell, taste, or feel a product.
ii. Sensory skills vary from person to person.
iii. All individuals need to be instructed on how to take a test.
iv. Not all individuals qualify for all tests, nor should they be expected to.
v. Subjects are rewarded for participation, not for correct scores.
vi. Skills once acquired are forgotten if not used on a regular basis.
vii. Skills can be overworked or fatigued.
viii. A subject’s performance can be influenced by numerous factors unrelated to the test or the product.
ix. All subject information should be treated in a confidential manner.
x. Employees should not be paid to participate in a sensory test.
xi. Test participation should always be on a volunteer basis.
xii. Subject safety is of paramount importance and should precede all other considerations.
xiii. Should not participate more than three or four times within a week, for not more than about 4
continuous weeks followed by 1 or 2 weeks of rest.
Experimental Design
• Experimental design is divided into two basic

structures, namely treatment structure and design
structure.
• Treatment structure as set of samples or treatments

that the client(s) selected to study in the specific
project – Control sample
• The design structure is defined by sensory specialists

when they group experimental units into blocks.
Experimental Design
• These two structures are linked by the randomization

performed by the sensory specialist prior to the study
and together they make up the experimental design of
the study.
• The sensory specialist should let treatment structure

dictate neither a poor design structure nor a favorite or
frequently used design structure affect the selection of
treatments.
Design Structures
A. Completely Randomized Design (CRD)
• In this design all the samples are randomly assigned to all the panelists.
• This technique works quite well in situations where the number of
samples is small and all samples can be evaluated by all panelists in a
single session.
• CRD is the ideal design for a central location consumer test where each
panelist evaluates each sample.
• For example, in a mall intercept, test panelists are asked to express their
degree of liking for each of four cola products. Each panelist receives the
four colas in a randomly assigned sequence.
• CRD designs also include random assignment of
products to people where each individual only sees one
product. These so-called consumer monadic tests are
common in consumer field studies.
• These are also called between-groups comparisons

since there are different groups of people evaluating
each product.
B. Randomized Complete Block Design (RCBD)
• In a randomized complete block design each treatment (usually
samples) is randomly assigned to each unit (usually panelists)
within each block (often sessions).
• This design is frequently used when trained analytical panelists

cannot evaluate all samples in replicate in a single session.
• In this case the best solution is to have each panelist evaluate all
samples in a single session and then have them return for a
subsequent session to re-evaluate all the samples.
• An example is a descriptive analysis study of six ice creams
made with fat replacers.
• In a single session the panelists can only evaluate six

samples. However, the samples should be evaluated in
triplicate. The panelists must attend three sessions to
evaluate all the samples in triplicate.
• In this study the blocks are the sessions and the six samples
are randomized across those panelists within each block.
C. Balanced Incomplete Block (BIB) Design
• Incomplete block designs are used when there are too many
treatments in the experiment for the panelists to judge all
samples in a single session (block).
• In this case the panelists evaluate subsets of samples in

individual sessions.
• The objective may be for each panelist to ultimately evaluate

all samples often in replicate or it may be that panelists only
see a subset of samples.
Treatment
Structures
• Kemp, S. E., Hollowood, T. & Hort, J. 2009. Sensory Evaluation: A Practical
Handbook. United Kingdom: Wiley-Blackwell
• Lawless, H.T. & Heymann, H. 2010. Sensory Evaluation of Food: Principles and
Practices. 2nd Ed. New York: Springer.

Planning Your Sensory Project

Uploaded by

Copyright:

Available Formats

Planning Your Sensory Project

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Planning Your Sensory Project

Uploaded by

Copyright:

Available Formats

PLAN YOUR

1. Setting the goals and objectives

• It is imperative that the potential outcomes of different methodologies

• When selecting an appropriate methodology to satisfy the desired

• When the test objective includes some aspects of performance,

• Financial constraints have to be considered in any test design.

• In studies that require several test elements, e.g. consumer tests,

5. Setting action standards

• Experimental design is divided into two basic

• Treatment structure as set of samples or treatments

• The design structure is defined by sensory specialists

• These two structures are linked by the randomization

• The sensory specialist should let treatment structure

A. Completely Randomized Design (CRD)

• These are also called between-groups comparisons

• This design is frequently used when trained analytical panelists

• In a single session the panelists can only evaluate six

• In this case the panelists evaluate subsets of samples in

• The objective may be for each panelist to ultimately evaluate

You might also like