‫اﻟﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ ﻟﻸﻏﺬﻳﺔ‬

‫‪1‬‬
2
3
 Introduction
• Sensory evaluation is commonly used in
quality control and product development of
food.
• Sensory evaluation is made by the senses of
taste, smell, vision , touch and hearing
• When the food is eaten, the complex
sensation that results from the interaction of
our senses is used to measure the food
quality

4
 Objective &Subjective methods
• Food quality can be measured either objectively
by using instruments like determination of
protein using Kjeldahl method, or objectively by
using people in the sensory evaluation method
• The importance of sensory analysis has grown
more and more inspite of the availability and
widespread of highly sensitive measuring
instruments such as HPLC, GC, AAS …etc

5
 Comparing instruments with sensory
tests
• Presently, food analyst are certain that optimal
measurement of food quality can be achieved
by coordinating between instrumental and
sensory evaluation
• In the instrumental analysis, there is a limit
where no signal appears such as 1 ppm
• While human senses which can be called
biological detectors may still perceive an odor
or taste which has a concentration less than 1
ppm

6
 Comparing instruments with sensory
tests
• The instrument can analyse only one
component while human senses can give a total
impression of aroma, taste and texture
• Accordingly, sensory analysis is very important
and equals if not superior to the instrumental
methods in the field of food product
development

7
 History

• Modern sensory analysis started only before 70

years in the Scandinavian countries (1940)
• Ten years later (1950), the European countries
used the sensory analysis in their food firms
• By 1970, sensory analysis developed into a
science. It is now being taught in many European
and American universities. University of
California at Davis as example offers a degree
program in sensory analysis
• Many food firms in the west offer industrial
courses in sensory analysis for their employees
8
 Establishing of sensory groups
• Establishing of scientific sensory groups started only
before about 15 years (1999).
• The 1st one was established in UK by The Institute of
Food Science &Technology.
• There are special conditions to be a member in
• such sensory groups.
• These sensory groups aim at dissemination of food
sensory evaluation culture through workshops, seminars
and conferences
• Fortunately, the Jordanian Sensory Group is being
established these days and it is expected to start its
activities very soon.

9
 Sensory groups

10
 Sensory groups

11
 Sensory groups

12
 ‫اﻟﺠﻤﻌﻴﺔ اﻷردﻧﻴﺔ ﻟﻠﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ ﻟﻸﻏﺬﻳﺔ‬

‫‪13‬‬
14
 ISO and Sensory evaluation
• The International Standard Organization (ISO)
started before about 20 years, standardization
of the sensory testing methods and a
significant work has been achieved.
• Good and reliable sensory results can be
obtained and analyzed statistically only if exact
and standard scientific sensory testing methods
are applied.

15
 • Terms related to sensory analysis:
• Panelists: persons who test the food and
judge it is named panel, panelist, assessors
or judge. Panelist may be one person or
several hundred depending on the type of
the sensory evaluation method.
• Testing Area: A special room or lab should
be designed for carrying the sensory tests.
This room should have partitions to
separate panelists and has suitable stools,
proper lighting, openings and a source of
water.
• The proper testing time was found to be
later morning or mid afternoon.

16
 Sensory laboratory

17
 Sensory laboratory

18
 Sensory terms
• Sample preparation for sensory evaluation :
• Kitchen: a well equipped kitchen is
necessary to prepare food samples for
sensory evaluation.
• Number of samples to be given, time or
duration between offering samples, soaking
or using water…etc. are important factors
that should be taken into consideration.
• Selecting the proper panel test and in this
respect, we have many tests where each one
suits a special purpose, among the different
panel tests we have :
• Uniformity in samples is important,
19
• Serving temperature should be selected properly
according to the type of food. For beverage a
temperature of 10-15 ºC is suitable while for ice
cream a temperature of -1 ºC is recommended.
In hot food like hamburger a temperature of 60
ºC is adviced.
• Containers:colorless or white , identical
containers should be used for each sample to be
tested. Disposable paper or plastic containers
are preferable.
• Quantity of sample, in the case of difference test,
this quantity is 16 ml for liquid or about 28 g for
solid , and in preference test the quantity is
20
doubled.
 Sensory terms
• Number of sample to be tested by panelist, it is
is variable and depends on the type of the food
to be tested and the experience of panels.
However, a preliminary test should be
conducted to determine how many samples can
be given to the panelists to evaluate them
properly.
• Coding which means giving a code or a
number to the samples to be tested is important
and should not affect the results. It is not
recommended to give numbers like 1, 2 or A, B ;
but it is better to code the samples with a three
digits numbers.

21
 Sensory terms
• Rinsing is also a step in sensory testing, it is
recommended to rinse your mouth with worm
water after testing fatty foods; in cheddar
cheese it is adviced to rinse with water , then to
wait 5 minutes before testing the next sample .
• Information about the sample: little information
should be given about the sample to be tested ,
otherwise the decision of the panels will be
affected by these information. For this reason
the persons involved in the preparation of the
tested samples should not be included in the
test.
22
 Sensory terms
• The panelists can be grouped to
 trained panelists,
 semi trained panelists and
 untrained panelists.
• In case of trained panelists a number of 4-5 is enough for
the test,
• while for semi trained a number of 10-20 is preferable and
• for untrained person, the larger the best.
• Panelist should be of good health and should stop
smoking, eating and drinking before 30 minutes of the
test. The panelists should be trained.
 Sensory Quality Attributes

• Food quality which can be detected by our

senses can be divided to 3 main factors:
• Appearance factors
• Textural factors
• Flavor factors
• Appearance factor can be judged by sight and
touch senses. They include size, shape, pattern,
wholeness, damage and defects.

24
 Sensory Quality Attributes

• Appearance factor also includes color, glossy,

transparency, turbidity and consistency (gel, flow,
viscosity and spread)
• Textural factors include the food properties
which can be judged by both mouth feel and
hand feel.
• Mouth feel properties include chewiness,
fibrousness, gritteness. Mealiness and stickiness.
• Hand feel properties include the softness,
firmness and juiceness.
25
Sensory evaluation and quality of food

• Sensory evaluation is considered one of the

most important tools used in measuring food
quality.
• There are some quality parameters that can not
be measured accurately by instruments and the
role of instruments is complementary to sensory
evaluation.
• Examples for such situation are the different
tastes and odors.
• Accordingly, we will focus on some quality
26 parameters that have to be measured sensorial.
 Sensory terms
• There are a special sheets or questionnaire for each
sensory test, the results should be tabled and analyzed
statistically to find if there is a significant difference or
not.
• Factors affecting sensory evaluation:
• expectation error (more information.
• Logical error (yellow color in potato)
• Suggestions.
• Contrast effect (poor or good sample).
• Stimulation error (other variables).
• Halo effect (more than one factor) .
• Motivation (importance).
• Positional (bias i.e. very simple different and selection
the middle sample).

27
 Sensory Quality Attributes
• Flavor factors quality parameters can be judged
by the two senses taste and smell.
• Under taste, we can find the terms salty, sour,
bitter, and sweet.
• Under odor , we can find the terms acid, burnt
and fragrant.
• Under the off-flavors we can find the terms over-
cooked, stale, enzymatic, chemical and
contaminated.

28
 Sensory Quality Attributes
• Flavor factors (taste and odor) are
subjective and difficult to be accurately
measured.
• It is possible to find hundreds of
descriptive terms regarding flavor
depending on the type of food.

29
 Complexity of flavor factor
• Flavor in food is very complex,
• an example is coffee flavor where more than
600 compounds responsible for coffee flavour
have been separated and identified
• and inspite of the great progress in the analytical
instrumentation in this field like the use of gas
chromatography techniques in identifying the
flavor components,
• flavor of food still a complex subject and not yet
fully understood, specially with regard to the
response of buds on the tongue to the different
tastes.
30
 Flavor factors are subjective
• Flavor factors include taste and odor (smell, aroma) and
can be detected by either tongue or nose. Taste and
odor are largely subjective and as a result hardly to be
accurately measured.
• Subjective: can be described as sweet… etc.
• Objective: can be measured and take a value as in ash,
protein…etc.
• Since flavor factors are subjective, there is a large
possibility for difference of opinion regarding evaluation.
However this is expected due to:
• People differ in their sensitivity to detect taste and odor.
• If they detect them correctly, people differ in their
preference.

31
 Effect of color and texture on the flavor
• Another thing which is important here is the
effect of color and texture on the flavor of the
food. During evaluation of flavor, color, and
texture often interfere with the evaluation of the
flavor.
• As an example flavor of chery and strawberry is
associated by people with the red color. Also
people frequently judge the thicker food (texture)
as richer in flavor.

32
 Methods for measuring flavor

• Panel Tests or Taste Panels

• Measuring Flavor:
• Volatile material by GC
• Flavor contributing substances by
chemical and physical means as: sugar,
salt and acid.

33
 Measuring flavor of food
• All these are research quality tools, but when it
comes to the consumer quality acceptance, the
above chemical, physical, and instrumental
methods, can not work. The only means is to
measure the quality using people.
• Measuring flavor of food:
• We have to use the panel tests; in conducting
panel tests, many points should be considered
such as the panelists or judges: their number,
trained or semi-trained or untrained; testing room:
the light, color of tables, chairs, ventilation and
so on.
34
 Methods for Sensory Evaluation
• Many methods for sensory evaluation are
available but the food quality control or product
developer should be familiar with the advantage
and disadvantage of each method and should
select the method which properly suit the
property to tested or examined in the food.
• Sensory evaluation methods can be classified to
3 groups as follow:
• Difference test.
• Preference tests.
• Descriptive tests.

35
 Methods for Sensory Evaluation
• Difference tests include the following methods:
– Triangle test
– Duo-Trio test
_ Simple paired comparison test.
– Scheffe paired comparison test
– Multiple paired comparison test.
– Ranking.
– Scoring.
– Ratio-scaling.
• Each of the above methods is applied for certain
evaluation test and has its own method of data
36
analysis.
 Methods for Sensory Evaluation

• Descriptive Test:
• Flavor profile test.
• Texture profile test.
• Preference Tests:
• Paired comparison and Multiple compression
preference tests.
• Ranking
• Hedonic scale test

37
 Choosing the proper panel test

• - Hedonic scale tests

• - Triangle tests.
• - Multiple comparison test.
• - Ranking test: taste the following food samples
and arrange them 1st, 2nd. 3rd and so on
according to your preference.
• Collecting sheets, tabulating, statistical analysis
and giving conclusions.
• These tests are applicable not only for
evaluation of flavor factor but also for other food
quality factors.

38
39
40
41
42
 1st lab
Recognition test for the four basic tastes

43
44
45
46
47
48
49
 Recognition test for the four basic
tastes
 Our senses of odor or smell and taste are as
important as our senses of vision and hearing regarding
the test sensation.
 Tongue and the hard palate are of interest for such
sensation.
 The four basic tastes are sweet and it is sensed at the
tip of the tongue salty or saline are sensed at the tip and
edge of the tongue, the third taste is the sour and it is
sensed at the edge of the tongue. The fourth taste is the
bitter and it is sensed at the back of the tongue .

50
51
52
 • In this lab we will practice the four basic
tastes.
• Sucrose soln will be used at two conc. i.e
0.4 and 0.6% to indicate the sweet taste.
• Sodium chloride will be used for salty taste,
and tow conc. will be used i.e 0.08 and
0.16% .
• Three conc. of citric acid (0.02 , 0.03 , and
0.04%) .will be used for identifying the sour
taste.
• Bitter taste will be identified using quinine
sulphate two conc. will be used i.e 5 and 10
ppm.
• Another soln will be given as a neutralizing
53
agent, it will be water at 40c.
 Preparation of stock and working solutions

54
 Samples Preparation

55
 Water Quality

56
 Glassware and cleanliness

57
 Coding of Samples

58
 Preparation of stock and working
solns .
• Trained personal,
• weighing and quantitative transference
• volume of the sample
• (test soln II)
• Sweet taste (sucrose) : 20g / 100ml (20%) stock soln .
(0.4%): 10ml from stock ------ 500ml (test soln I)
• (0.6%): 15ml from stock ---- 500ml (test soln II)

• Salty taste (sodium chloride):10g/100ml(10%)stock soln

0.08 4ml / 500ml (test soln I)
• 0.15 7.5ml / 500ml

59
 Preparation of stock and working
solns .
• Sour taste (citric acid) : 1 g/100 ml (1 %) stock soln
• 0.02 10ml / 500ml (test soln I)
• 0.03 15ml / 500ml (test soln II)
• 0.04 20ml / 500ml (test soln III)
•
• Bitter taste (quinine sulphate) : 50mg /1000ml (0.005%) stock
soln
• 5 ppm 50ml /500ml (test soln I)
• 6 ppm 120ml / 1000ml (test soln II)
• As you remember we can use three methods for preparing soln, the
1st is % w/w (weight / weight), the 2nd is % w /v and the 3nd is % v/
v
•

60
61
 Taste testing technique

62
 Taste testing technique

63
64
65
66
67
 Results presentation and
discussion

68
 Test sheet filling

69
70
 3rd Lecture: Odour recognition test

• We can smell while breathinng, but we

perceive odours more intensely by sniffing.
• The sniffing technique, however, should
not be applied to all odorants as some of
them have a pungent component causing
pain.
• We can taste an odorous substance on
the tongue (without taking it into the
mouth) by the von Skramlik test.

71
• Odours can be memorized as one
memorizes a foreign language.
• A panel leader with
comprehensive technical
knowledge and skill will be able to
provide many memorization aids
so that the panelist’s learning
time can be considerably
72shortened.
 Objectives of Odour recognition test

• The main objective for such test is to train

our odour memory.
• Odour can be memorized as one
memorizes foreign language.
• It is recommended not to offer more than
3 odours per session to beginners.

73
• Some facts have to be taught to
beginners.
• In many cases we do not
recognize odours which we
encounter daily, because we do
not pay attention to them.
• These fact always surprises the
participants during their first
odour recognition tests.
74
• Do not make excessive demands
upon beginners. It is better to present
two or three odours per session over
a longer period of time than to give
up after 20 odorants in one session.
• No one is necessarily a good test
subject because he is able to
recognize a number of odours in the
first sessions.
• These may have been familiar to him
75
but not to others
• Of more importance is the ability
to memorize unknown odours
and to be able to characterize
them in detail.
• This ability is later required in
descriptive tests (on food quality,
off-notes etc.).
• In order to avoid fatigue, different
(not similar) odours should be
presented in one session.
76
• In the literature, it is always
recommended that ‘panellists
should have good health, and
should not suffer from chronic
colds’.
• No hints are given, however, as to
improve health.
• For many persons the causes for
chronic colds and/or a blocked
nose are unknown.
77
• Medical literature explains that chronic
colds may have their origin in suppurated
dead teeth and/or root residues.
• As the nerves have been killed and
removed, inflammation and suppuration
remain undetected because they do not
cause pain.
• Chronic sinusitis can also cause constant
colds.
• After following medical advice, better
results have been obtained from such
panellists.
78
• The difference between single and
complex odours is not only
dependent upon the odorants but
also upon the background of the
candidate.
• Eau de Cologne, for example, is
described as a ‘single odour’ by many
beginners.
• The trained perfumer, however,
analyses the single character notes
of this ‘complex odour’.,

79
• A number of odorants have, in addition to
their odour, a pungent component which
causes pain.
• Some people are excessively sensitive to
pungency which makes it impossible for
them to perceive the odour of the sample.
• In this case, one drop of the odorant is
added to 50 or 100 ml of water.
• In smelling the aqueous solution (Or
suspension) the pain causing pungency
becomes so weak that the odour is
perceived without difficulty.

80
• When tasting this solution the odour can
be perceived more intensely as it
becomes more volatile when warmed in
the mouth.
• This technique should always be used by
persons who are more taste-sensitive than
odor-sensitive.
• Once the odour has been perceived in this
way it will be more easily perceived by
smelling at a later time

81
 Anatomy of the nose

• The nose has the task of smelling as well as

warming and filtering of the air we breath before
it passes to the lungs.
• There are three horizontal folds in the nose
where air passes through during breathing (area
a) , (area b) and area “R”.
• Area “a” for normal breathing while “b” for
sniffing.
• The upper part of the nose “R” which is called
olfaction region is for odour sensation (fig 1).
82
 Human Nose Passages and Folds

83
 Physiology and smelling technique

• The inspired air does not stream through

the upper part of the nose (R area) in
ordinary breathing ,
• to make air reach to the “R” area sniffing is
needed for this purpose.
• Sniffing for three times is satisfactory for
odour sensation otherwise fatigue is
encountered .

84
 Von Skramlik test

• The main objective of this test is to prove that

the presence of odorants in air is not sufficient to
cause odour sensation.
• It is done by closing the nose by fingers and by
inhaling and exhaling through the mouth for an
odorant.
• After practicing by the mouth for at least 3 times,
while the bottle of the odorant is close to the lips,
the mouth is closed and the nose is opened.
• At this time the odour sensation takes place.

85
 Von Skramlik test

86
 Reccommendations
• Large no of odors may cause physical and
mental exhaustion for several days.
• Large no of odorants as well as the order
of their presentation my cause odor
fatigue.
• It is advised to smell sweet odor like
vanillin after having smelled fruity one like
lemon oil or vice versa.
• By such technique we can avoid odor
87
fatigue.
 Reccommendations

• It is recommend to exclude subjects suffering

for cold during odor recognition test.
• Sometimes, it is difficult to exclude them, so
they can have special treatment to cancel the
cold effect. This can be achieved by special
medical treatment (teeth repair) or by taking
Japanese medicinal herb oil.
• Odorants having pain causing component should
not be inhaled in a concentrated form but diluted

88
 Samples preparation and presentation

• Most odorant are presented either in 10%

or 1% soln.
• Solvent might be water or ethanol.
• Solid sample might be ground and diluted
with sugar such as cinnamon.

89
• Sample solutions are presented either on
smelling strips or on soaked cotton wool.
Filter paper might also be used to such
purposes.
• Time intervals between sample
preparation and odor sensation is variable.
• Some should be prepared on the day of
sensation others before one week such as
slightly rancid coconut flakes.

90
 Taste and odor blindness
• Sometime we may find subjects that are
bitter taste blinders.
• This is very dangerous since consuming
of 7 and 60 raw bitter almond kernels
are fatal for children and adults
respectively.
• The same apply also for odor blind
subject.

91
• Benzaldehyde which is formed due to
hydrolysis of amygdaline (a glycoside in
almonds and apricot) is not poisonous
and smell similarly to better almond note
• but hydrocynamic acid formed due to the
hydrolysis of amygdaline is poisonous
• and this explains why odor or taste
blindness may be sometimes very
dangerous.

92
 2 nd lab : odour recognition test
• Our today lab will concentrate on the odour
recognition test as well as the Von Scremlik test.
• A special sheet will be distributed for you and
some odor samples.
• Yon have to practice the odor sensation and to
fill the sheet with the correct answer.
• Don’t forget to describe the odor you examine.

93
• Before odor recognition test you have to try the
Von Scramlik test.
• A discussion will follow the test to see our
correct or wrong answers.
• Results might be presented on the blackboard.
• In case of significant incorrect answers we have
to repeat the tests more than one time with more
concentration and sometimes with concentrated
odorants.

94
95
96
 Aroma recognition test

 Anatomy of the nose :

 In the last week during demonstrating of odour
recognition test it was mentioned that there are
3 folds in the nose i.e.
 (a) , for normal breathing ,
 (b) for sniffing and
 (R) which is called olfaction region for odor
sensation.
• The aroma of a product can be perceived both
by smelling and tasting for example, coffee
aroma

97
  Further derails about the anatomy of the nose
 The nose has two openings; the nostrils are
the anterior nares and the openings at the
back of the throat are the posterior nares or
choana.
 Regarding aroma, its perception is achieved
by two means i.e.
 by the nostril or what is called the anterior
nares (smelling) and
 by the posterior nares or what is called
choana (tasting).

98
 Physiology of aroma perception

• As shown in Fig. 6.3, two paths lead to the regio

olfactoria (R); externally through the nostrils (a),
and internally through the choana (b).
• The odorous constituents of a substance are
frequently more strongly perceived by tasting
than by smelling.
• When taking the sample into the mouth, it
becomes warmed and the odorous constituents
become more volatile.
• This is the reason why, in odour recognition tests,
tasting instead of smelling was suggested for
people having difficulties when smelling.
99
100
 Demonstration of the physiology of Aroma
perception

In order to distinguish between aroma

perceived by smell and aroma perceived
by taste, the nose has to be blocked.
The panel leader demonstrates the
exercise without reference to the test
substance (cinnamon sugar).
He gives the following instructions:
 Hold beaker Z with the sample at a
distance of about 20 cm.
101
 Remove the aluminium foil cover, but do not look
into the beaker and do not smell the sample!
 Take one spoonful of the sample, but hold it
away from you.
 Block your nose with the other hand and keep it
blocked while you ‘take the sample into your
mouth (see Fig. 6.4).
 Chew the sample with your mouth open (in order
not to suffocate).
 Analyse the aroma (flavour).
 You will perceive only ‘sweet’.

102
103
 Slurping a special test
technique
• As swallowing samples is not wholesome , a
special technique was devcloped by coffe and
tea testers called slurping.
• It replaces the sensory action of swallowing
• In swallowing , the aroma filled air is
pressed into the regio olfactoria via the posterior
nears .
• This also occures in slurping but takes more
effort.
• For some people , it may take a long
time to learn the correct technique of slurping.
104
• The following has to be donc vcy qLnckly . close the mouth and keep it c ose w n C YOU open t e nose. n y now w
ien you he able to perceive the aroma of... (the participants shou 1(1 write down or tell what they prcccive).
This exercise is so impressive that it should never be left out
The story of the apple and tJI OfliOll
in the literature , it is. stated that commmutcd apples and onions cannot be differentiated from one another
when tasted with eyes and nose closed also ,there is many similar pairs of samples
I —red wine and unsweetened black coflc.
2—a raw apple and raw potato.
3-a tomato and an orange.
4-d ii uteci raspberry syrup and sweetened milk
Samles which cannot be differentiated when tasting withcyes and
nose closed, it is important that the samples differ merely in their odorous constituents . if one milk samples is
sweetened with sugar ( 42 . 75 gm
sucrose to 1 litre whole milk ) and the other with vanillin sugar ( 42 . 75 gin vanillin sugar (5 small packages) to I
litre whole milk) no
differentiation is possiple . another suitable comparison is scalded milk versus burned milk . (it is rcommended
that the samples be pourccI
through a sieve befor taste-testing.)
*sarnpi iresentation
masking the colour with red or with sodium vapour light the colour differences are often not completely masked
coloured glasses or plstic backers offer another way to masking coloures, colouring of samples is another wa’ of
masking colour and is in common use in the confectionery industry ( colouring of candies ) and mostly wc we
use red light for colour masking.
Also ,that important to keep glasses odou rfree , by fill it with 3-5 % acetic acid and soak over night then rinsing
with clear water.

105
 The story of apple and onion
 Apple and onion can’t be differentiated from
each other when are properly prepared i.e.
peeled and ground twice (the texture should not
be a variable).
 The work should be in pairs, the question to be
asked : Is it apple or onion?
 The samples should be given randomly and their
no. should be 4 pairs at least.
 The eyes and nose should be closed.
 Such test will teach us the importance of the
olfaction region

106
 Aroma recognition test
• Different foods are given to the students
to get experience with various types of
food aroma.
• Fruit and vegetable juices are widely used
for such purposes.
• Juices used are either natural or diluted.
• Color differences have to be masked by
colored light, by using colored beakers or
by coloring the samples.
107
 • Aroma recognition tests for juices will be
conducted using special forms.
• One of the sheets will be for aroma recognition
by smelling, while the other by tasting.
• Each form also include aroma description.
• Another sheet or form containing the results
will be distributed to evaluate your results.
• In case of incorrect results for any juice, you
have to test again , to get familiar with the
aroma of that juice.

108
 Off-notes
Off-notes such as coated or rancid flavor
are of vital important with regard to the
aroma recognition in the food industry.
Milk is highly sensitive food and widely
used for aquiring the off notes.

109
• Two test groups will be organized on the off
notes using milk.
• The 1st to identify the bitter , watery , cooked ,
scorched and sour aroma.
 The 2nd test to identify or demonstrate the
metallic, cardboard , rancid , cooked aroma or
flavor.
 All the off- notes tests will be practiced using the
smelling and tasting technique.
 Don’t forget not swallow the sample , only chew
it and describe , then recognize the aroma.
 It is better not to see the sample.

110
111
112
 ‫اﻟﻄﻌﻮم اﻷﺧﺮى‬
Additional Sensation

 Tastes perceived during tasting can be

classified to basic tastes and additional
tastes or sensation.
 As you know, the basic tastes include, sour,
sweet bitter and salty.
 The additional sensations include
temperature sensations, pain sensations.
Touch sensations, kinesthetic sensations,
sound sensations and color sensations.
 Temperature, pain and touch sensations are
called cuteneous sensations.

113
 I. Temperature Sensation
• ‫ اﻹﺣﺴﺎس ﺑﺎﻟﻄﻌﻢ اﻟﺪاﻓﻲء‬: ‫أوﻻ‬
 Some foods and beverages are preferably consumed at
specific temperatures. Tea and coffee and soups are
consumed hot or warm at about 60 C.
 Other foods like soft drinks are consumed cold at about
5 – 10 C, while ice-cream is consumed very cold at 0 C.
 Cold and warm or hot sensations are not caused only by
tasting hot or cold foods and drinks, but there are some
substances that can cause such sensations.
 There are coolness receptors and heat receptors on the
tongue where consuming some food sensitizes these
receptors.
 As you see in the table below

114
• Ia. Warm sensations and mixed warm and
pain sensations
• Warm Sensation
• a. warm (15 % Ethanol)
• b. Warm (Powdered Cinamon)
• c. Burning (Warm+Pain) (Powred Ginger)
• d. Neutrilizing Agent ( Butter)
 Table 7.1. Exercise 1a. Warm sensations perceived while tasting and
mixed sensations of warm and pain

116
• Ib. Cooling, cold and very cold sensations
• Cold Sensation
• Cool Sensation (Cocoa butter or cocoa fat)
• Cold Sensation (Menthol)
• Very Cold Sensation (Eucalyptus Oil)

 Cooling sensations are due to the stimulation of

coolness receptors or sensitizing them through
absorption of health by such substances.
 It is important to open your mouth and draw in some air
to feel the cold sensation.
 It is important to take a neutralizing agent after
completion of each excersize and before starting a
second one.

117
 II. Pain Sensation

Tasting or smelling of some

substances may cause pain sensation.
 This pain sensation ranges between
very slight pain or tart to a high pain or
burning sensations.
 Many terms are used to describe the
pain sensations and their increasing
intensity and as appear in the following
table.
118
 Pain Sensation

• ‫ اﻹﺣﺴﺎس ﺑﺎﻹﻟﻢ‬: ‫ﺛﺎﻧﻴﺎ‬

IIa. Pain sensations ranging from tingle to pain
• Tingle and slight pain (Carbonated beverage)

II.b. Pain sensations ranging from astringent to buckery

• Astringent (0.3 % Alum)
• Pronounced astringent (Undiluted lemon juice) or
(0.4 % alum)
• Pukery (0.5 % alum)
• Neutrilizing agent: (Milk chocolate)

119
120
 • II.c Pain sensations ranging from
sharp to burning
• Pungent or pronounced sharp
• ( 5 % Acetic acid or vinegar)
• Burning ( Pepper)
• Neutralizing agent (Butter)

121
 III. Touch sensations

• It includes pressure and tactile sensations. It is

difficult to differentiate between these two
sensations and only trained panalists can do that.
• If you touch the back of your hand with your
finger tip without moving back and forth you will
have the pressure sensation.
• Since there this no movement, pressure
sensation is called static sensation.
• Tactile sensation can be felt as before but with
moving the tip of the finger back and forth. So
tactile sensation is called dynamic sensation.

122
 Pressure and tactile sensations

• Sensation of different particle sizes are included

under the tactile sensation.
• It is clear from the table bellow that small,
medium and large particle size can be perceived
while tasting.
• Pressure and tactile sensations as appear from
the table cover the pairs of opposite
• Sensation.
• Touch sensations also include the fat and oil
feeling or coating.

123
 • ‫اﻹﺣﺴﺎس ﺑﺤﺠﻢ اﻟﺤﺒﻴﺒﺎت‬:‫ﺛﺎﻟﺜﺎ‬
• III.a.Pressure and tactile sensations: Particle size
• Sandy (Small particle size)
• (Calcium carbonate)
• gritty (Medium particle size)
• (Semolina)
• Grainy (Large particle size)
• (Rice)

124
 • ‫اﻹﺣﺴﺎس ﺑﺎﻟﻄﻌﻢ اﻟﺪﻫﻨﻲ أو اﻟﺰﻳﺘﻲ‬
• II. III.b.Pressure and tactile sensations: Oil
feeling or coating
• Smooth (Butter)
• Oily:Oil feeling or coating
• (Olive oil)

125
126
• III.c.Pressure and tactile sensations: Opposite
pairs
• Mouthfilling (Fullcream UHT milk)
• Watery or flat (Skim UHT milk)
• (‫اﻟﻨﻌﻮﻣﺔ واﻟﺼﻼﺑﺔ)اﻟﺤﺴﻴﺔ اﻟﺤﺮﻛﻴﺔ‬: ‫راﺑﻌﺎ‬
• Kinesthetic sensation
• Soft (White bread)
• Medium hard (Zwieback)
• Very hard (Hard candies)

127
 IV.Kinesthetic sensations

• Kinesthetic sensations are perceived while

tasting. In this sensation we use our muscle or
power so it is sometimes called sensation of
resistance.
• When we chew tough meat we will experience
strong resistance while when we chew tender
meat we experience little resistance.
• Many terms are used to describe the kinesthetic
sensations as appears from table IVa.

128
 Kinesthetic sensations

• ‫اﻟﻬﺸﺎﺷﻴﺔ واﻟﻠﺰوﺟﺔ واﻹﻧﻘﺼﺎﻓﻴﺔ واﻟﻐﻀﺎﺿﺔ‬: ‫ﺧﺎﻣﺴﺎ‬

• IV.a.Kinesthetic sensations: Hardness
• Hard candies----------- very hard
• Zwieback -------------- medium hard
• White bread ----------- soft

129
• IV.b.Kinesthetic sensations: Brittle
• Very brittle (Potato chips or corn flakes)
• Crackly (Cookies or biscuit)

• IV.c.Kinesthetic sensations: Thin and thick Soups

• IV.d.Kinesthetic sensations: sticky and slippery

• Sticky----------- dry whole milk powder
• Slippery -------- raw egg white

130
 V. Sound sensation

• During the consumption of some hard foods, we

get a sound sensation.
• Mainly three terms are used to describe the
sound sensation as appear in table V.
• Table V. Sound sensation
• a.Crackly--------------- biscuit or cookies
• b.Crunchy-------------- Chips
• c. Moist Crispy------- Pickles (slightly moist crisp)
• Raddish (pronounced moist crisp)
• Cookies (dry crispy)

131
132
 ‫إﺧﺘﺒﺎرات اﻟﻌﺘﺒﺔ‬
Threshold tests
• The recognition tests for the four basic
tests are considered qualitative tests.
While the thresholds tests are quantitative
tests.
• In the thresholds tests we aim at
determining the concentration at which the
particular basic tests can be recognized.
• We are measuring the sensitivity of the test
subject or determining his threshold.

133
 Threshold types
• Threshold types: there are four threshold types:
• The first threshold types is called the detection
or stimulus threshold.
• In this type of threshold tests, it is reached when
the solution of the basic taste is given in such a
concentration where it tastes different than the
water and the basic taste is unrecognizable.

134
 The second threshold type
• The second threshold type is the recognition
threshold.
• Here the solution of the basic taste is given in
such a concentration where the taste can be
correctly recognized when.
• We reach to this point we can say that we
reached to the recognition threshold, and the
concentration value is called the threshold value
such as 0.15% NaCl.

135
 Third threshold type
• The third threshold type is the difference
threshold.
• It is defined as the concentration of the solution
of the particular taste where a difference in taste
can be tested or recognized.
• Trained panel can detect or recognize difference
in taste at 10% concentration 1%, 1.1%, 1.2%,
…etc. of NaCl while for beginners a
concentration of 30% is needed to detect the
difference: 1%, 1.3%, 1.6% of NaCl…etc.
136
 The fourth threshold type
• The fourth threshold type is the
terminal threshold.
• It is defined as the concentration of the
basic taste solution where above it, an
increase in the concentration can't be
detected or recognized.

137
 Information value of threshold
tests
• A low recognition threshold obtained with
an aqueous series does_not necessarily
correspond to a high sensitivity to foods.
• Therefore threshold tests should not play
too important a role in a, sensory course.
• We regard them as a basic exercise but
shall pay more attention to tests with
foods as we do not consume only
aqueous solutions in our daily life.
 Combination of threshold tests with intensity
exercises

• It is suggested that we first detetmine the

approximate threshold by presenting a series
with large graduations and high concentrations
(see Tables 9.1—9.4, part a).
• For each concentration the intensity has to be
defined.
• We will discover that we are not able to
differentiate between small concentration
differences because the human senses are
better able to differentiate qualities (e.g.
different odours) than quantities (different
intensities).
• Therefore, scales should not be used
(especially in quality rating) with any finer
graduations than can be differentiated.
• For the determination of the exact
threshold value concentration series with
finer graduations are presented (see
Tables 9.1—9.4, part b).
• The results will become even more exact
when difference tests are applied (as will
be shown in Chapter 10).
 Points to be considered in
threshold tests

• preparation of the concentration series

– Preparation of stock solution:
it is advised when preparation of solution of
increasing intensity or concentration to
prepare first a stock solution and to prepare a
working solution with the different intensity or
concentration from this stock solution.

141
 Preparation of stock and working
solutions
• For sweet basic taste the concentration of stock
solution is 20% or 20g sucrose/100 ml; the
concentration is wt/vol.
• For salty basic taste, the concentration of the
stock solution is 10% NaCl or 10g/100 ml.
• For the sour taste, the concentration of the stock
solution is 1% citric acid
• For the bitter taste the stock solution is either
0.05% caffeine or quinine hydrochloride.

142
 Preparation of stock and working solutions

• To prepare solution such as 0.05% w/v

quinine hydrochloride.
• We weigh 0.125g which is equal to 125 mg
very precisely using a high sensitive
balance,
• dissolve it in little quantity of H2O then
transfer quantitatively to 250 ml volumetric
flask and add water till we reach to the
mark.

143
 Water used in preparation

Quality of water is very important for

sensory evaluation of foods in general and
for threshold tests in particular.
The following types of H2O are nor
accepted in threshold tests:
o chlorinated water,
o Too hard water,

144
 Water used in preparation
o heavily deionized water,
o demineralized water ,
o distilled water.
o The types of H2O may exhibit
unacceptable odor, mask sweetness,
introduce bitterness or susceptible to
microbial contamination like
demineralized water.
145
 Preparation of H2O for threshold purposes

• Many researches recommended to

prepare the H2O for threshold purposes
in special way.
• The method includes the boiling of fresh
H2O for 10 minutes in an uncovered
enamel pot or glass beaker at least 15
cm in diameter.
• The obtained boiled water is cooled
then filtered and the first portion is
discarded.
146
 Preparation of the test or working solution

• It is advised to prepare two kinds of

concentration series of testing solution.
• One with the upper concentration for
beginners and the second with lower
concentration for trained persons.
• This mean that the graduation between the
upper concentration solution is large like
0.10, 0.15, 0.20 %... etc.
• While a finer graduation is found in the
lower concentration solution for trained
persons like .0.10, 0.12, 0.14…etc.
147
 Number of solutions to be prepared for test
solutions

• Many investigators advise to have

10-12 testing solutions with a gradual
concentration.
• For a sweet taste, an initial
concentration of 0.00 and final
concentration 1% (0.0, 0.05, 0.1,
0.2…1%) is prepared.
• For salty taste, the initial concentration
is 0.00 and the final or the tenth
solution is 0.2%.
148
• In sour taste, the initial concentration is
0.00 and the tenth solution concentration
is o.o35
• For bitter taste the final concentration is
0.0003 in case
of quinine HCL and 0.030% for caffeine.
• The quantity of testing solution is 500 ml
for 15 testers and 1000 ml for 30
testers.
 Coding of solutions and beakers

• It is better to have the same coding on beakers

and volumetric flasks having the prepared test
solutions from 2 digital numbers like from 11 to
20 or 21 to 30 or 31 to 40 and so on.
• The beakers also should have the code number
of panelist between two brackets such as 11(2)
for panelist number 2 and solution with the code
11.
• Since it is advised to have two threshold tests at
a sessions, twenty beakers of two different
colors is preferred.
• The quantity of tested sample is about 30 ml.
150
 • The beakers should be arranged in a manner to
save time and avoid disturbancy.
• In case one taste is tested only, the beakers
should be in one line with increasing number...
11.12.13…etc. in case of two series ( Sweet
and Sour) the sweet series be in front and the
sour behind.
• With regard to the time of the test, it is
suggested not to have or to run the threshold
test on empty stomach. Suitable time is one
hour after breakfast or one hour after lunch.

151
 Explanation of the test

• First the mouth is rinsed with the control water to

be familiar with its taste.
• Testing of samples is started by beaker 11 as
example. Then followed by 12 and so on.
• It is very Important to note that swallowing of
the sample is not allowed.
• Only the sample moved around in the mouth in
such away to come in contact with all part of the
tongue.
• Also do not repeat testing any sample.

152
 • After rinsing the mouth with the sample
number 11 we have to ask the following
question: does it tastes sweet, bitter, sour or
salty?
• If the answer is it likes water a (0) is marked
on the sheet.
• If the answer was that the sample tastes
different than water but the taste quality can't
be recognized, then we write a question
mark(?) on the sheet.
• The test is continued with samples 13, 14
and so on.
153
• There is another column in the test sheet
specific for intensity of the taste.
• Numerical value have to be given to
intensities rather than descriptive words i.
e. sour 1, sour 2, sour 3 instead of very
weak sour, weak sour, and sour. Stronger
concentration will be given number 4 and
the strongest number 5.
155
 Transparency number 1 shows the test sheet and how it
can be filled correctly.

156
 Factors affecting threshold test results

– Insensitivity for the 3 basic tastes; sour, bitter

and salt:
– sour insensitivity to observed with people who
lacks gastric acid;
– bitter insensitivity is innate and can only be
slightly improved by training.
– Salty insensitivity cab be caused by teeth
filling on having denture plate.
– This insensitivity can be improved by
changing food habits such as adding less or
little salt to food.
157
 – Oversensitivity for sweet taste: it is observed
in people with too much gastric acid.
– Sex may affect the threshold test, women
were found to be more sensitive to sweet and
salty while men are more sensitive to sour.
– Age also affect the threshold test. It is
observed that sensitivity for sweet and sour
decrease with age.
– Alcohol was found to affect widely the
sensation of the four basic tastes. After Two
hours of taking only 25ml ethanol, the trained
panel subject was not able to differentiate
between the four basic tastes.

158
• - presentation of results and discussion:
will be during the exercises.

159
 ‫اﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ‬
‫‪ .1‬أﻫﻤﻴﺔ اﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ‬
‫ﻻ ﻳﺨﻔﻰ ﻋﻠﻰ أﺣﺪ اﻟﺪور اﻟﻤﻬﻢ واﻷﺳﺎﺳﻲ ﻟﻠﺘﺤﻠﻴﻞ •‬
‫اﻹﺣﺼﺎﺋﻲ ﻓﻲ ﻣﺠﺎل اﻟﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ إذ ذﻫﺐ‬
‫اﻟﺒﻌﺾ اﻟﻰ أﺑﻌﺪ ﻣﻦ ذﻟﻚ وﻗﺎﻟﻮا ﺑﺄﻧﻪ ﻻ ﺗﻘﻴﻴﻢ‬
‫ﺣﺴﻲ ﺑﺪون ﺗﺤﻠﻴﻞ إﺣﺼﺎﺋﻲ‪ ،‬أي ﻻ ﻗﻴﻤﺔ ﻟﻠﻨﺘﺎﺋﺞ‬
‫اﻟﻤﺘﺤﺼﻞ ﻋﻠﻴﻬﺎ ﻣﻦ اﻻاﺧﺘﺒﺎرات اﻟﺤﺴﻴﺔ إذا ﻟﻢ‬
‫ﻳﺆﺧﺬ ﻓﻲ اﻟﺤﺴﺒﺎن ﺗﺤﻠﻴﻞ اﻟﻨﺘﺎﺋﺞ إﺣﺼﺎﺋﻴﺎ‪.‬‬
‫وﻳﻨﺴﺤﺐ ذﻟﻚ ﻋﻠﻰ اﻷﻣﻮر اﻟﺨﺎﺻﺔ ﺑﺘﻄﻮﻳﺮ •‬
‫اﻟﻤﻨﺘﺠﺎت اﻟﻐﺬاﺋﻴﺔ أو ﻋﻠﻰ أﻣﻮر ﺿﺒﻂ ﺟﻮدة‬
‫اﻷﻏﺬﻳﺔ‪.‬‬
‫وﻣﻦ ﻫﻨﺎ ﻳﻼﺣﻆ أن ﺟﻤﻴﻊ ﻃﺮق اﻟﺘﺤﻠﻴﻞ اﻟﺤﺴﻲ •‬
‫اﻟﺘﻲ ﺗﻢ ﺗﻄﻮﻳﺮﻫﺎ واﻋﺘﻤﺪت ﻛﻤﻮاﺻﻔﺔ دوﻟﻴﺔ ﺗﺸﻤﻞ‬
‫ﺿﻤﻦ ﺑﻨﻮدﻫﺎ ﻋﻠﻰ ﺟﺰء ﺧﺎص ﺑﺎﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ‬
‫ﻟﻠﻨﺘﺎﺋﺞ اﻟﻤﺘﺤﺼﻞ ﻋﻠﻴﻬﺎ ﺑﺘﻠﻚ اﻟﻄﺮﻳﻘﺔ‪.‬‬
‫‪160‬‬
 ‫‪ .2‬اﻟﻤﺮاﺣﻞ اﻟﺘﻲ ﺗﻤﺮ ﺑﻬﺎ ﻋﻤﻠﻴﺔ اﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ‬
‫* ﻣﺮﺣﻠﺔ ﺗﺼﻤﻴﻢ اﻟﺘﺠﺮﺑﺔ إﺣﺼﺎﺋﻴﺎ وﻳﺸﻤﻞ ذﻟﻚ ﻋﺪد •‬
‫اﻟﻌﻴﻨﺎت‪ ,‬ﻋﺪد اﻟﻤﻘﻴﻤﻴﻦ‪ ،‬اﻟﻄﺮﻳﻘﺔ اﻟﺘﻲ ﺳﺘﺴﺘﺨﺪم ﻓﻲ‬
‫اﻟﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ‪ ،‬اﺳﺘﻤﺎرة اﻹﺧﺘﺒﺎر اﻟﺤﺴﻲ ‪ .....‬اﻟﺦ‪.‬‬
‫*ﻣﺮﺣﻠﺔ ﺗﻘﺪﻳﻢ وﻓﺤﺺ وﺗﻠﺨﻴﺺ ﻧﺘﺎﺋﺞ اﻟﺘﺤﻠﻴﻞ اﻟﺤﺴﻲ •‬
‫وﻗﺪ ﻳﺸﻤﻞ ذﻟﻚ ﻋﻤﻠﻴﺎت ﺟﺪوﻟﺔ اﻟﺒﻴﺎﻧﺎت اﻟﺤﺴﻴﺔ وﺗﺤﻮﻳﻠﻬﺎ‬
‫اﻟﻰ ﻣﻨﺤﻨﻴﺎت ورﺳﻮﻣﺎت‪.‬‬
‫*ﻣﺮﺣﻠﺔ اﻟﺘﺤﻠﻴﻞ ﺑﺎﺳﺘﺨﺪام إﺣﺪى اﻟﻄﺮق اﻹﺣﺼﺎﺋﻴﺔ •‬
‫اﻟﻤﻨﺎﺳﺒﺔ‬
‫*ﺗﺸﻤﻞ اﻟﻤﺮﺣﻠﺔ اﻟﺮاﺑﻌﺔ واﻷﺧﻴﺮة اﻟﻮﺻﻮل اﻟﻰ •‬
‫اﻻﺳﺘﻨﺘﺎﺟﺎت ﺑﻨﺎء ﻋﻠﻰ ﻧﺘﺎﺋﺞ اﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ‪ .‬ﻓﻤﺜﻼ ﻟﻮ‬
‫ﻛﺎن ﻋﻨﺪﻧﺎ ﻧﻮﻋﻴﻦ ﻣﻦ اﻟﺒﺴﻜﻮت وﺗﻢ ﺗﻘﻴﻴﻤﻬﻤﺎ ﺣﺴﻴﺎ‬
‫وﺣﻠﻠﺖ اﻟﻨﺘﺎﺋﺞ اﺣﺼﺎﺋﻴﺎ ﻓﻔﻲ اﻟﻤﺮﺣﻠﺔ اﻟﺮاﺑﻌﺔ ﻳﺘﻢ‬
‫‪161‬‬ ‫اﻻﺳﺘﻨﺘﺎج واﺗﺨﺎذ اﻟﻘﺮار اﻟﻤﺘﻌﻠﻖ ﺑﺄي ﻧﻮع ﻣﻦ اﻟﺒﺴﻜﻮت ﺗﻢ‬
‫ﺗﻔﻀﻴﻠﻪ ﻣﻦ ﻗﺒﻞ اﻟﻤﺴﺘﻬﻠﻚ‪.‬‬
 ‫ ﺑﻌﺾ اﻟﻤﺼﻄﻠﺤﺎت اﻟﻤﻬﻤﺔ ﻓﻲ ﻣﺠﺎل اﻟﺘﺤﻠﻴﻞ‬.3
‫اﻹﺣﺼﺎﺋﻲ ﻟﻠﺒﻴﺎﻧﺎت اﻟﺤﺴﻴﺔ‬
• Data value and data set
• Observations
• Variables
• Null hypothesis and alternative hypothesis
• Types of error and how to measure
• Variance
• Probability and significance
• One tailed and two tailed experiments

162
163
 STATISTICAL ANALYSIS OF THE RESULTS (DATA)

• (a)Fundamentals of statistics
• *In order to facilitate the
understanding of statistics some
basic terms are explained as follows:

164
 • Null hypothesis
• *This is a very important term in
statistics.
• For our purposes it will be the
assumption that a result is based
solely on chance.
• This would mean in our example that
a test subject is simply guessing and
is therefore completely unqualified as
a panel member.

165
 • We now want to find out what
requirements must be fulfilled by a
test subject in order for him to be
considered qualified or,
• which is the point at which the null
hypothesis can be rejected.

166
 • Alternative hypothesis
• *While, in the problem described, the null
hypothesis states that the result is due to
chance or guessing,
• the alternative hypothesis states that it is not
due to chance;
• the test subject is able to differentiate
correctly.

167
 • Level of error
• *In the example with five testing pairs, the
null hypothesis (result by chance or guessing)
has been rejected.
• It might, however, occur that a test subject
has passed this qualification test merely by
guessing.
• In this case, the null hypothesis has been
unjustifiably rejected, and is called level of
error (α).

168
• Error of the first and second kind
• *When the null hypothesis has been unjustifiably
rejected, this is an error of the first kind.
• This means, in the case of the second type of
paired difference test (with same and different
samples in each pair),
• that differences have been found where they
have not existed (pair with same samples).

169
• *When the alternative hypothesis has been
unjustifiably rejected, this is an error of the
second kind.
• This means that no differences have been
found although the samples were different.
• The latter error easily occurs when unqualified
test subjects are kept on the panel.
• Therefore, it requires that only the best test
subjects be selected.

170
• (b) Statistical analysis of paired difference tests
• Table 10.1.6 reveals whether differences between two
samples have been found simply by chance or
whether they can be considered significant.
• This table, which refers only to one-tailed tests,
contains data for three levels of significance or three
limits for α, the level of error.
• In the case of a 5% level of error, the stronger tasting
sample can be correctly identified simply by chance
(the null hypothesis has been unjustifiably rejected), in
one out of 20 cases.

171
• *In this example, the statistical significance
(given in percentage) is S = 100 - α = 95%.
• This means that the result is significant, or is
secure at the 5% level of significance,
• which is marked on the head of the table as α
= 0.05 or with one asterisk (*).

172
 • *In the case of a 1% level of error, there
is a danger that in one out of 100 cases
the stronger tasting sample has been
correctly identified simply by chance.
• In this case, the statistical significance
(given in percentage) is S = 100 – α=
99%.
• The result is then highly significant or
secure at the 1% level of significance,
• which is marked on the head of the table
as α = 0.01 or with two asterisks (**).
173
• *In the case of a 0.1% level of error, there
is a danger that in one out of 1000 cases
the stronger tasting sample has been
correctly identified simply by chance.
• In this case, the statistical significance
(given in percentage) is S = 100 - α =
99.9%.
• The result is then very highly significant
or secure at the 0.1% level of significance,
• Which is marked on the head of the table
as
• α = 0.001 or with three asterisks(***).
174
 • The statistical significance increases
as the level of error decreases.
• Let us analyse statistically the results
of the three exercises which we have
carried out.
• We start by counting the number of
judgements, which is equal to the
number of test subjects multiplied by
the number of pairs:
• 20 (number of test subjects) X 4
(number of pairs) = 80 (number of
175
judgements)
176
177
 ‫اﻟﻌﺸﻮاﺋﻴﺔ ﻓﻲ ﺗﻘﺪﻳﻢ اﻟﻌﻴﻨﺎت ﻟﻠﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ‬
‫ﻳﻮﺟﺪ أﻛﺜﺮ ﻣﻦ رأي أو اﺗﺠﺎه ﻓﻴﻤﺎ ﻳﺘﻌﻠﻖ ﺑﺘﺮﻗﻴﻢ أو ﺗﺮﻣﻴﺰ •‬
‫اﻟﻌﻴﻨﺎت ﻷﻏﺮاض اﻟﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ ﻛﻤﺎ أن ﻟﻨﻮع اﻹﺧﺘﺒﺎر اﻟﺤﺴﻲ‬
‫ﺗﺄﺛﻴﺮ ﻋﻠﻰ ذﻟﻚ‪.‬‬
‫ﻓﻬﻨﺎك ﺑﻌﺾ اﻹﺧﺘﺒﺎرات اﻟﺘﻲ ﻳﺘﻢ ﻓﻴﻬﺎ اﺳﺘﻌﻤﺎل أﻋﺪاد ﻣﻦ •‬
‫رﻗﻤﻴﻦ ﻟﺘﺮﻣﻴﺰﻫﺎ ﻣﺜﻞ ‪ ... 23 ،22 :21‬اﻟﺦ‪ .‬وﻓﻲ ﺣﺎﻟﺔ‬
‫اﺧﺘﺒﺎرات اﻟﻌﺘﺒﺔ ﺗﺴﺘﺨﺪم اﻟﺤﺮوف ﻓﻲ ﻋﻤﻠﻴﺎت اﻟﺘﺮﻗﻴﻢ‪ :‬أ‪ ،‬ب‪،‬‬
‫ج‪ ،‬د‪ ،‬ﻫـ‪ ،‬و‪ ....‬اﻟﺦ‪.‬‬
‫وﻣﻊ ذﻟﻚ ﻓﺈن أﻏﻠﺐ اﻟﻌﺎﻣﻠﻴﻦ ﻗﻲ ﻣﺠﺎل اﻟﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ •‬
‫ﻳﻔﻀﻠﻮن اﺳﺘﻌﻤﺎل أﻋﺪاد ﻣﻦ ﺛﻼﺛﺔ أرﻗﺎم ﻟﺘﺮﻣﻴﺰ اﻟﻌﻴﻨﺎت‪.‬‬
‫وﻣﻤﺎ ﻫﻮ ﺟﺪﻳﺮ ﺑﺎﻟﺬﻛﺮ أن ﻃﺮﻳﻘﺔ ﺗﻘﺪﻳﻢ اﻟﻌﻴﻨﺎت ﻳﺠﺐ أن •‬
‫ﻳﺨﻀﻊ ﻟﻠﻌﺸﻮاﺋﻴﺔ وﻣﻦ اﻟﻀﺮورة ﺑﻤﻜﺎن اﺳﺘﺨﺪام اﻟﺠﺪاول‬
‫اﻟﺨﺎﺻﺔ ﺑﺬﻟﻚ واﻟﻤﺴﻤﺎة ‪.Random digit‬‬
‫‪178‬‬
 ‫اﻟﺠﺪول اﻟﺨﺎص ﺑﺎﻟﻌﺸﻮاﺋﻴﺔ‪Random digit table‬‬

‫‪179‬‬
 ‫اﻟﻄﺮق اﻟﻤﺴﺘﺨﺪﻣﺔ ﻓﻲ اﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ ﻟﻠﺒﻴﺎﻧﺎت‬
‫اﻟﺤﺴﻴﺔ‬

‫ﺗﺘﻮﻓﺮ ﺛﻼﺛﺔ ﻃﺮق ﻟﺘﺤﻠﻴﻞ اﻟﺒﻴﺎﻧﺎت اﻟﺤﺴﻴﺔ إﺣﺼﺎﺋﻴﺎ وﻛﻤﺎ ﻳﻠﻲ‪• :‬‬
‫اﺳﺘﺨﺪام ﺟﺪاول ﺧﺎﺻﺔ ﺑﺎﻟﺘﺤﻠﻴﻞ اﻹﺣﺼﺎﺋﻲ وﻟﻜﻞ ﻃﺮﻳﻘﺔ •‬
‫ﺗﻘﻴﻴﻢ ﺣﺴﻲ اﻟﺠﺪاول اﻟﺨﺎﺻﺔ ﺑﻬﺎ ﻓﺘﺘﻮﻓﺮ ﺟﺪاول ﺧﺎﺻﺔ‬
‫ﺑﻄﺮﻳﻘﺔ اﻟﺘﺮﺗﻴﺐ ) ‪ (Ranking test‬وأﺧﺮى ﻹﺧﺘﺒﺎرات ﻣﻘﺎرﻧﺔ اﻷزواج ) ‪Paired‬‬
‫‪ (comparison test‬وﻫﻜﺬا‪.‬‬
‫اﺳﺘﺨﺪام ﺗﺤﻠﻴﻞ اﻟﺘﺒﺎﻳﻦ وﺳﻴﺘﻢ اﻟﺘﻌﺮض ﻟﻪ ﺑﺸﻲء ﻣﻦ اﻟﺘﻔﺼﻴﻞ •‬
‫ﻻﺣﻘﺎ‬
‫اﺳﺘﺨﺪام ﻃﺮق ﺧﺎﺻﺔ وﻣﻨﻬﺎ اﺳﺘﺨﺪام ﺑﺮاﻣﺞ ﺣﺎﺳﻮﺑﻴﺔ ﺧﺎﺻﺔ •‬

‫‪180‬‬
 Analysis of variance

• Chewiness
• --------------------------------------------------------------------
• Brands
• --------------------------------------------------------------------
• Judges A B C D Total
• ---------------------------------------- ------------------------------- -
• 1 2.2 1.9 3.2 0.2 7.5
• 2 2.5 1.0 4.2 2.0 9.7
• 3 3.3 3.0 4.0 1.5 11.8
• 4 2.6 5.4 4.2 1.3 13.5
• 5 2.6 4.3 5.1 0.1 12.1
• 6 2.2 3.4 4.7 1.9 12.2
• 7 2.4 3.6 5.5 3.2 14.7
• 8 1.6 3.9 5.7 1.8 13.0
• --------------------------------------------------------------------------
• Total 19.4 26.5 36.6 12.0 94.5
• ----------------------------------------------------------------------------
• Mean 2.4 3.3 4.6 1.5

181
 Correction factor (CF) = 94.52/32
= 8930.25/32
= 279.07
Sum of squares, samples=(19.42 + 26.52 + 36.62 + 12.02)/8 – CF
=2562.17/8 – CF
=320.27 - 279.07
=41.20
Sum of squares, judges =(7.52 + 9.72 + 11.82 + ... + 13.02)/4 – CF
=1152.17/4 – CF
=288.04 - 279.07
=8.97
Sum of squares, total = (2.22 + 2.52 + ... + 1 .82 – CF
=347.09 - 279.07
=68.02

182
• Analysis of variance table
• ----------------------------------------------------------------------------------
• Source of df SS MS F
• Variation
• ----------------------------------------------------------------------------------
• Samples 341.20 13.73 16.1 5*
• Judges 7 8.971.281.51
• Error 21 17.85 0.85
• Total 31 68.02
• ---------------------------------------------------------------------------------

183
• The F value for samples is 16.15.
• According to Chart 3 in the Appendix,
page 65, if the F value exceeds 3.07 there
is a significant difference at the 5% level (*)
and if it exceeds 4.87 there is a significant
difference at the 1% level (**).
• There is a significant difference at the 1%
level for samples so the value is marked
with two asterisks. There is no significant
difference for judges.

184
• Many statistical tools are used to test the differences
between means and if these differences are significant
or not.
• Among these tests:
• Tukey's Test,
• Multiple range test, and
• Least significant test (LSD)
• LSD = t X SE
• The standard error of the mean is calculated by the
equation:
• SE = √0.85/8
• = √ 0.1062
• = 0.32

185
• The t value is taken from( t) table for 21
degrees of freedom and it is 2.08.
• Least significant difference=2.08 X 0.32 =0.67
• Sample scores:A B C D
• 19.4 26.5 36.6 12.0
• Sample means:2.4 3.3 4.6 1.5

186
• The sample means are arranged according to magnitude.
• C B A D
• 4.6 3.3 2.4 1.5
• Each mean is compared with the others to see if the
difference is 0.67 or more.
• C - D = 4.6 - 1.5 3.1 > 0.67
• C- A = 4.6 - 2.4 = 2.2 > 0.67
• C - B = 4.6 - 3.3 = 1.3 > 0.67 (least significant
difference)
• B - D = 3.3 - 1.5 = 1.8 >0.67
• B - A = 3.3 - 2.4 = 0.9 > 0.67
• A- D = 2.4 - 1.5 = 0.9 > 0.67

187
 • The results are shown using
letters to indicate differences:
C B A D
• 4.6a 3.3b 2.4c 1.5d
• Any two values not followed by the
same letter are significantly different
at the 5% level. Sample C is
significantly more chewy than all the
others. Sample B is more chewy than
sample D.
188
189
190
191
192
193
194
195
196
197
 ‫ ﻃﺮق‬: ‫ﻃﺮق اﻟﺘﻘﻴﻴﻢ اﻟﺤﺴﻲ‬
‫اﻻﺧﺘﻼف‬
Difference tests
• Objective of the method
• Difference tests are analytical methods.
The panel members act like an instrument
whose task is to analyse whether or not
there is a difference between two samples.
• As the differences are very small, these
tests are quite difficult.
• They require high sensitivity which,
however, can be acquired through training.
198
 Questioning

• The question ‘which is the odd sample' or

'which sample is stronger' is posed in
such a way that no hedonic answer can
be given.
• This is necessary in order to keep the
method strictly analytical.

199
 Special requirements to be met by the samples

• If only the odour, taste or flavour has to

be analysed, the samples must have the
same appearance, shape, temperature,
and consistency, and equal amounts
have to be presented.
• Otherwise bias might occur.

200
 Methods of Application
• Difference tests are used in
• product testing and
• panel training
• and panel selection.
• Not all five methods are, however, equally suitable.
• The five main difference test methods are:
• 1. The paired difference test
• II. The triangle test
• 111.The duo-trio test
• IV. Ranking tests
• V. Multiple Comparison Difference Test

201
 Difference threshold

• Last week the threshold terms were

explained to you;
• (1) stimulus or detection threshold,
• (2) recognition threshold,
• (3) terminal threshold.
• (4) The fourth is the difference
threshold. It is defined as that
concentration at which the difference is
202
recognized.
 Difference threshold
• Beginers receive samples with a 30%
difference (e.g. 1%, 1.3% sodium
chloride solution).
• With training and much concentration
during the test, a difference threshold
of 10% can be reached (e,g. l.0, 1.10 %
sodium chloride solution).

203
 Information value of difference
tests
• In contrast to tests with a concentration
series, difference tests can determine
whether or not a test subject is suited
for sensory analysis.
• A low difference threshold not only for
aqueous solutions for the four basic
tastes but mainly for foods is a quality
of a good sensory analyst.
204
 Special requirements

• Difference tests require a high degree of

concentration.
• Final panel selection should not be made
during training time, as some candidates may
learn slowly but show considerable
improvement with time and perseverance.
• Such improvement may not be observed with
people who learn easily by memorization.

205
 Number of tests and test subjects

• Managers often have difficulty concentrating

due to their busy schedules and are,
therefore, not suitable as permanent panel
members.
• Number of test subjects
• For difference tests, at least ten people
should be available, but panels with 20 or 30
members are better.
• Number of tests
• More important than the number of tests
carried out in a sensory laboratory is the
thorough planning and execution of sensory
tests.
206
 PAIRED DIFFERENCE TEST
1.1 EXPLANATION OF THE METHOD

• Directional paired difference test

• Several pairs are given to each panel
member.
• Each pair consists of a control sample K (e.g.
apple juice) and an analytical sample A (e.g.
apple juice with sugar added).
• Pairs differing only in concentrations can
also be presented (A = low, B = higher
concentration)
207
 • The objective of the method is to state
whether the two samples can be
differentiated.
• The question posed may be: which
sample in each pair is sweeter (saltier,
more sour, more fruity, more rancid,
more aromatic, more crisp, etc.).
• Only one question is posed in one test.

208
• Sample arrangement
• The position (left or right) of the analytical (or
the more concentrated) sample in each pair is
arranged randomly by consulting tables of
random digits. By changing the sample
arrangement from test to test, guessing is
avoided.
• Examples of sample arrangements are:
• KA AK AK BA AB AB
• AK or AK or KA BA AB BA
• AK KA AK BA AB AB
• KA KA KA BA BA AB

209
 Second type of the paired difference
test

• Pairs with the same and different samples are

presented.
• The question is posed whether the samples in
each pair are identical or different.
• Possible sample arrangements are:
AK, KA, KK, AA.
• In practice, this type of paired difference test is
used mainly for the training and checking of test
subjects.
210
 Exercise

• Comparison of Natural apple juice and

Sweetened apple juice
• The goal of this exercise is to become
familiar with the differences in sweet
intensity.
• It is, therefore, better suited for the
determination of the intensity differences.

211
 Number of samples and sample amount

• Number of samples
• Four pairs per single person, these are called a
'set'.
• Sample amount
• 30 ml per single sample.
• When using four pairs per test subject, 120 ml of
0.018% tartaric acid (sample A) and 120 ml of
0.021% tartaric acid (sample B) are needed.
• For 15 persons, about 2 L of each sample are
needed.

212
 Sample preparation

• Sample preparation
• As has been explained earlier.
• Tasting beakers
• Eight tasting beakers are required per person.
• The beakers are coded: 21, 22, 23, 24, 25, 26, 27,
28.
• It will be explained later (see triangle tests) why
we use these codes.
• The beakers are arranged in pairs, the pair with
the lowest numbers in front and pairs with higher
numbers behind.
213
 Sample preparation

• On the test sheet, the pair with the lowest

number is written on the first line and that with
the highest on the last line.
• This is frequently not done by beginners who
also place, for example, the beakers of all four
pairs in a row or mix up the left with the right
beaker and vice versa.
• Care must also be taken so that distribution
errors do not occur.

214
 Paired difference test questionnaire

• Test product: Apple juice Name:

• Test subject No.: 15 Date:
You will receive several pairs of apple juice samples. Please write
the pair with the lowest number on the first line and that with the
highest on the last line.
• Circle in each pair the number of the more sweet tasting sample.
Retasting is allowed.
Taste quality to be tested: Sweetness of apple juice
• Sample numbers Remark
• 21 22
• 23 24
• 25 26
• 27 28
 Table 10.I.3 - Paired difference test

• (Example of filled out test sheet)

Test product: Apple juice Name:
• Test subject No.: 15 Date:
You will receive several pairs of apple juice samples.
Please write the pair with the lowest number on the first
line and that with the highest on the last line.
• Circle in each pair the number of the more sweet tasting
sample. Retasting is allowed.
Taste quality to be tested: Sweetness of apple juice
• Sample numbers Remark
• 21 (22)
• (23) 24
• (25) 26
• 27 (28)

216
 Example of a sample arrangement

• According to Table 10.1.7, which uses random

digits, the samples have to be arranged as
follows:
• -----------------------------------------------------------------
Random digit Even/odd number Underline the Sample
codes
8 Even number Right code 21
22
1 Odd number Left code 23
24
5 Odd number Left code 25
217 26
 • This means that the more sweet apple juice
is poured into the beakers with the underlined
codes (22, 23, 25, 28) and the less sweet
apple juice into the remaining beakers (21,
24, 26, 27).
• Sample coding
• The same codes are used for all participants
of a course so that the results can be
discussed immediately after the test.
• In regular tests, however, each panel member
receives samples with different codes.

218
 Statistical analysis of the obtained results
In Paired difference test, the obtained results are statistically analyzed
using special table as below

219
220
 Difference tests:
2. TRIANGLE TEST

• 2.1. EXPLANATION OF THE METHOD

• (a) Simple (classical) and extended triangle
test
• Importance
• The triangle (or triangular test) is the most
widely used of all different tests.

221
 Objective of the method

• In the simple or classical triangle

test the only task is to find out
whether or not there is a difference
between two samples.
• In the extended triangle test, there
are additional tasks.

222
 Simple triangle test

• Three samples are presented, two are

alike, one is different. The question
posed is: "which is the odd sample?"
Either the control sample (K) or the
analytical sample (A) are presented in
duplicate, thus allowing six
combinations:
• AKK KAK KKA and
• KAA AKA AAK
• (Similar combinations are possible
when comparing two commercial
samples A and B).
223
• The samples can be arranged in a
triangle (hence the name), but are
usually presented in a row.
• Samples having an after-taste should
not be duplicated as this would make
the test too difficult

224
 Extended triangle test

• Additional tasks may be defining the

degree of difference (triangle-intensity
or triangle-directional test), and/or
characterizing the difference.
• When descriptions are too difficult,
noting the panel member's preference
(for either the duplicate or the odd
sample) may provide hints for product
development.
225
• More information will be obtained,
however, when the reasons for the
preference can be characterized in detail
by the panel members.
• Reasons for Primary emphasis on
exercises with the extended triangle test
• Panel results which only indicate a
significant difference between two
samples are of little value in industrial
practice.

226
 • In quality control and in research and
development further information is
required:
• How can the difference between the
samples be characterized?
• Was the preference for the one or the
other sample significant?

227
 Method Application

• The triangle test is of equal importance in

product testing and in panel training and
selection.
• Problems to be solved in product testing
are manifold.
• In industry, production may suddenly be
halted because an off-note has been
detected.
• What is its origin?

228
 • Has the addition of salt or a spice
component been forgotten in the
manufacture of cookies?
• Did oil from a machine come into
contact with the chocolate during its
liquid stage?
• Has a new packaging material
caused an odour transfer?
• What else might have caused an off-
note?

229
 • In research and development it is not
sufficient simply to note a difference
between samples.
• The kind of difference must be
characterized to provide sufficient
information for further work.
• Is the new composition as
successful as the former one?

230
 • The lack of raw materials due to war or
a poor harvest in the country of origin
might have forced a change in
composition.
• Or, a food law may no longer allow the
use of a successful ingredients in
flavour compositions.

231
 • The flavour chemist wants to know
which of his fractions has interesting
and typical flavour compounds to be
further investigated.
• Triangle tests are also important in
storage tests (what change has
occurred?)
• and in the comparison of competitive
products (is there a difference?

232
 • What are the characteristics?
• Which product is preferred?).
• Different manufacturing processes
may also cause aroma changes (e.g.,
spray drying vs. freeze drying).

233
 Simple (classical) triangle test

• Comparison of natural and artificial pepper oil

• Test product Pepper oil
Name Test subject No.: Date:
• You will receive a set with 4 pairs each having
three samples.
• Two of these samples are identical and the other
is different.
• Circle the number of the odd sample.
• Re-tasting is allowed.
• Sample numbers:Remarks:
• 32 34 36; 42 44 46; 52 54 56; 62 64 66
234
 Storage tests

• Storage tests are becoming

increasingly important due to the legal
requirement of stating "keeping time"
on product packages.
• Therefore, as many triangle tests as
possible should be made with stored
foods.

235
 • The storage time can be shortened by
keeping the samples (milk, fats, meat
products) at room temperature as
against refrigerated.
• For safety reasons, only odour tests
should be made.

236
 • Cookies, chips and extruded peanut
snacks are stored on trays in order to
expose a large surface to light and air.
• The optimal storage time has to be
determined in pre-tests with a small
group of test subjects.
• Attention must also be paid to texture
changes especially with crisp foods.

237
 (b) Samples

• Sample coding
• In the literature, it is recommended that 1, 2,
3 or a, b, c not be used as codes.
• According to psychologists there is a danger
of choosing sample 2 or sample b as the
odd one.
• Psychologists, therefore, have proposed
three-digit random numbers as codes.
• Bias for the sample numbered "2" has not
been observed either in our laboratory.

238
 • According to our observations, three-
digit numbers provide no guarantee
against bias.
• By presenting samples with codes such
as 738, 955 and 231 there was a
tendency to choose the sample with the
lowest code number (231) as the odd
one.
• We have solved this psychological
problem by using either even (2,4,6 or
12, 14, 16 or 22, 24, 26) or odd number
239
codes (1, 3, 5 or 15, 17,19 or 23, 25, 27
• The number of samples per session
depends upon
 the test material, and
 the fatigue it may cause.
 The degree of difficulty also has to be
considered.
• During our intensive courses, we were
forced to present a large number of
samples within one and half days.

240
 • Participants had come a long distance and
had to gain as much practical experience as
possible in a short time.
• Outside of these courses, however, we present
only one or two triangles in one session.
• In special cases, panel members have to come
to a second session. This procedure provides
optimal testing conditions.

241
 Sample arrangement:

• Only during courses is the same sample

arrangement used for all participants in order to
save time during the subsequent discussion of
results.
• In our daily tests for research, development, etc.,
samples are presented in random order to each
test subject.
• This can be done even with the same sample
codes.

242
• On the master sheet, the A-samples of
each triangle are underlined.
• On the master sheet, the beaker with the
A-sample is underlined.
• Two numbers are underlined in the same
triangle when the A-sample occurs in
duplicate.

243
 c) Number of sessions

• As already mentioned, the thorough planning and

execution of sensory tests is more important
than the number of tests carried out.
• Inexperienced panel leaders frequently call for
too many sessions which are ill planned.
• They prepare the samples in a rush, are not
ready when the panel members arrive, state at
the last moment colour and/or temperature
differences, and prepare insufficient sample
amounts.

244
• Such panel leaders should not be
surprised when the test subjects lose
interest in sensory analysis and find all
kinds of excuses not to participate in
further sessions.
• The experienced panel leader can
convince a project leader that hurried
tests provide only questionable results.

245
• He always pre-tests to determine optimal
requirements for sample preparation and
presentation, and to select the appropriate
test method.
• This enables him to obtain maximum
information with a minimum of panel
sessions.
• This is very important for the
requirements of the practice, is cost
saving, and provides the most useful
results for research.
246
 (d) Test sheets

• An example of a correctly filled out test

sheet for the simple (classical) triangle
test is shown in Table 10.11.3 (see also
Fig. 10.II.1).
• Table 10.11.10 shows a test sheet for the
extended triangle test with several tasks.
• The test sheets may differ in some ways
from those reported in the literature.

247
• Industrial practice requires streamlining.
• Therefore, the code number of the odd
sample is not written in a special column.
• It is sufficient to circle it as was done in
the paired difference test (Table 10.1.3).

248
 Simple (classical) triangle test
• Comparison of two brands of black currant
nectar
• In case of colour differences, this test has
to be done under red light or the samples
have to be presented in dark-blue ATL-
glasses. Two triangles are presented. The
following codes are used:
• 43 45 47 (first triangle) BBA
• 54 56 58 (second triangle) ABB
• The number of the odd sample is
underlined on the master sheet.
• The sample amount is 30 ml per beaker.
Care must be taken to have sufficient
sample material
249
 Table 10.II.4 - Simple (classical) triangle test

• Comparison of Labeneh
• Test product: Labenah Name:
• Test subject No.: Date:
• You will receive a set of one triangle with three
samples. Two of these samples are identical and
the other is different. Circle the number of the
odd sample. Retasting is allowed.
• Sample numbers:Remarks:
• 32 34 36

250
 Testing sequence

• The samples in each triangle are tested from left

to right, thus guaranteeing that random order has
been followed.
• This refers especially to daily tests with a
random sample arrangement, but should also be
followed during courses.
• Retasting is allowed
• This can, however, cause fatigue if it is done too often.
• Frequently, the first impression is the correct one.
• Triangle tests require much concentration which
is acquired through training.

252
 Hints for beginners

• It may be helpful to describe the odour, taste and

flavour of each sample before continuing to the
next one.
• This allows neutralizing between the samples
which is not otherwise possible if the subject
has a poor memory.
• Early sample description training is a great
advantage for more difficult test methods
(flavour profile test). Notes can be made on the
back of the test sheet.
• As a neutralizing agent for the black currant
253
nectar, water at 35-40 "C is sufficient.
 Presentation of results at the blackboard

• To provide a summary of the group's

achievement, the number of responses are
tallied as shown below.
• First triangle
• 43 45 47
• / / //// ////
• Only at the very end is the code of the odd
sample circled by the panel leader, as shown
below:
• 43 45 (47)
• / / //// ////
• The differentiation is highly significant and can
254
be recognized without statistical tables.
 Presentation of the results at the notice board

• The results of all triangle tests are tabulated

separately from those of the paired test in a
record form as shown in Table 10.11.5.
• This provides a quick visual survey of the
achievements of each test subject.
• Differentiation is even better when the plus signs
are written in red and the minus signs in black.
• The sensitivity of single test subjects towards
the various test materials can also be easily
recognized.

255
 • People who are not sensitive to rancidity
should not be chosen for tests with fats,
whereas rancid-sensitive test subjects
should be selected to participate
especially in arbitration analyses.
• A separate record form with the test
subjects' numbers and names has to be
prepared for the panel leader.
• Test subject numbers may change.

256
• For staff planning, a record of each test
subject's achievement over a period of
several years is important.
• In our fragrance and flavour
manufacturing industry, only test subjects
with good sensory abilities over a period
of about three years have a chance to be
appointed to positions requiring sensory
sensitivity (as perfumers, flavourists,
sensory analysts, etc.).

257
 (c) Extended triangle tests with several tasks

• In industrial practice, it is as important to

characterize the single sample and the
duplicate as it is to determine the
sample preference.
• Sometimes characterization of the
samples is not consistent enough to
provide sufficient information for further
product development.

258
• It may, however, be valuable to state that
the preference goes in one direction (e.g,
when comparing a new flavouring
composition with an old one which can no
longer be used because of changes in
food law regulations).
• It may also be valuable to state about
equal preference for the two flavouring
compositions (e.g. in the case of imitating
a competitor's flavouring composition).

259
260
261
Multiple Comparison Difference Test
Questionare for difference test
 III. DUO-TRIO TEST

• EXPLANATION OF THE METHOD

• The due-trio test is so called because it is
intermediate between the duo (paired) and the
trio (triangle) test.
• The test subject receives one sample marked
control (K) and one or several pairs of samples.
Each pair contains one control sample and one
analytical sample (A) in random order.
• The question posed is: 'Which is the
different sample in each pair?'
• It can also be asked: "Which sample in
each pair is identical to the control?
• The latter question is frequently preferred
as it seems easier to answer.
• To avoid confusion it is recommended
that the same question be used for all
tests.

264
• Sample arrangement
• There are two possible sample combinations:
• K-KA or K-AK
• For several pairs consult Table 10.1.7 (ten thousand
random digits, p. 200).
• Example of possible sample arrangements
• K AK K KA K KA K AK
• KA AK KA AK
• KA AK KA KA
• A K AK KA K A, etc.

265
 Application

• Because the duo-trio test requires

little retasting it is used with samples
having a sharp or burning taste (pepper)
and/or a lingering aftertaste (bitterness)
.
• As p (probability) = 1/2, more tests are
required for statistically significant
results than in the triangle test with p =
1/3.

266
 • This method is also advantageous for the
recognition of off-notes.
• When the sample with the off-note is
presented in the labelled control beaker,
training time is considerably shortened.

267
 Exercise
• Two commercial brands of bitter
chocolate are presented.
• To avoid differences in appearance the
samples are cut into small pieces or even
grated. A red light can also be used.
• Sample presentation is shown in Fig.
10.111.1 and a correctly filled out test
sheet in Table 10.111.1.
 Re-tasting

• Re-tasting only the sample identical to the

labelled control not only saves time but
also avoids fatigue, especially when more
than two pairs are presented (e.g. fruit
juices).
• Errors which frequently occur in testing
the first pairs will easily become obvious.

269
• Suitable neutralizing agents
• for confectionery products are rose-hip
tea and/or water at 35-40 "C.
• Cider vinegar with honey can also be used
as follows: 2 tea spoonsful of cider
vinegar and 1-2 tea spoonsful of honey are
put in a drinking glass and filled with cold
or warm water.
• Dissolve the honey well.
 test sheet

• Test product: White cheese Name:

• Test subject No.: Date:
• You will receive a control sample and several pairs of
• samples.
• In each pair circle the number of the sample which
tastes
• different to the control.
• Re-tasting is allowed.

• K 41 42
• 43 44
 Table 10.1II.1 -Duo-trio test

• (Example of a correctly filled out test sheet)

• Test product: Bitter chocolate Name:
• Test subject No.: Date:
• You will receive a control sample and several pairs of
• samples.
• In each pair circle the number of the sample which
tastes
• different to the control.
• Re-tasting is allowed.

• K 41 (42)
• (43) 44

272
 Statistical analysis

• For statistical analysis of the duo-trio

test consult Table 10.11.12 (p. 219) for
significance in paired tests.
• Both of these tests (comparison of two
commercial brands of chocolate, and
of natural and artificial pepper oil) are
two-tailed as only the test result
reveals whether or not the samples are
different.
273
 Statistical analysis

274
 Tabulation of test results

• In Chapter 10.II., a triangle test with pepper

oil (natural vs artificial) was described. We
also used the duo-trio test for
methodological comparison.
• As shown in Table 10.III.2, significant results
were not achieved in the triangle test but
very highly significant results were achieved
in the duo-trio test.
• The labelled control not only facilitated the
differentiation but also made frequent re-
tasting of the burning pepper-oil samples

275
 III. Ranking tests

• 1. EXPLANATION OF THE METHOD

• Importance and application
• When using difference tests (paired,
triangle, duo-trio test), only two samples
can be compared at the same time.
• In practice, however, a series of
samples (commercial products, flavour
• compositions) may have to be judged

277
 • and their quality pre-selected (which
samples are better, which medium, and
which are poor quality).
• This pre-selection helps to save time, and
material (if there is only a small quantity of
the competitor's product available or if the
yield of the production is very small).
• For such purposes, ranking test is used. The
ranking test is also widely applied in
hedonic tests.

278
 • Testing task
• A series of samples has to be ranked
either:
• (a) in increasing intensity of a
property (sweetness, saltiness,
(b) aroma/flavour, rancidity);
• (c)for quality (comparison of
competitive products, flavours, etc.);

279
 • (d) for hedonic properties (like/dislike,
Preference, acceptability).
• This method has the advantage that
more than two samples can be
compared at the same time (in
difference tests, only two samples can
be compared at the same time).
• A disadvantage can be inexact results in
the case of small differences and large
variations.
280
 • Sequence of exercises
• Never start with quality ranking,
especially with beginners.
• Prerequisites for such ranking are not
only the ability to differentiate well,
but also extensive product knowledge.
• Quality ranking is done preferably in an
advanced course after completion of
exercises on quality description and
scoring.

281
 • Beginners should start by ranking the
intensity of the various senses (odour, taste,
colour., texture).
• Ranking the degree of preference is best
done with a larger group of laymen, as
trained panellists are already too
knowledgeable to be representative of the
consumer’s likes and dislikes.
• We shall, however, conduct such a test to
get acquainted with the test sheets and the
statistical analysis of the data.

282
 RANKING TEST IN ORDER OF INCREASING INTENSITY
OF TASTE (Four basic tastes)

• (a) With sucrose solutions

• Concentrations: 0.1, 0.4, 0.5, 0.8, 0.8, 1.0, 1.5, 2.0
g/100 ml solution.
• Sample preparation: see Table 9.1, p. 165.
• Sample amount: 30 ml per person.
• Marking the beakers: A, B, C, D, E, F, G, H. (The
beakers for the recognition test for the four basic
tastes can be re-used).
• For sample coding see Table 1l.2, P.256.

283
 • Beaker arrangement: in alphabetical order.
• Test instructions: The test subjects are told
that they will receive a sucrose concentration
series in random order.
• The task is to rank the samples in increasing
order of intensity.
• Two samples, however, will have the same
concentration and must be so noted.

284
 Details of the test

• To avoid taste fatigue from too frequent

retasting, it is recommended that the
approximate intensity of each sample first be
determined and written on the test sheet (Table
11.1)
• *After completion of the series, the final intensity
is to be determined.
• It will become superfluous to re-taste samples
with extreme intensities (weakest and strongest
sample).
• Samples with questionable intensity difference
are compared preferably in pairs (Fig.11.1).
• Finally, the samples are ranked in increasing
order of intensity and the result written on the
lower line of the test sheet (Table 11.1).
• Do not forget to mark the samples with the
same concentration.

286
 Table 11.1 - Ranking test (Example of a filled out test sheet)

• Name: Test subject No.: Date:

• You will receive, in random order, samples which differ in intensity for one property. In this test the
property is: Sweetness
• Start by determining the approximate intensity of each sample and applying the given intensity scale.
Frequent retasting will not be necessary. Note that two samples have the same intensity.
• Finally, rank the samples in order of increasing intensity. Retest only samples with questionable
intensity. Mark the duplicate sample.
•
• Sample Intensity Intensity scale
• codes (preliminary) final
• A (2) 3 1= very weak
• B 2 2 = weak
• C (5) 4 3 =medium strong
• D (2) 1 4 = strong
• E (4?) 4 5 = very strong
F (1) 2
• G (4+) 5
• H 3
• Final ranking
• Sample codes
B D F A H E C G
weakest strongest
Intensity intensity
Duplicate A + H

288
 Ranking test
• Name: Test subject No.: Date:
• You will receive, in random order, samples which differ in intensity for one property. In this test the
property is: Sweetness
• Start by determining the approximate intensity of each sample and applying the given intensity scale.
Frequent retasting will not be necessary. Note that two samples have the same intensity.
• Finally, rank the samples in order of increasing intensity. Retest only samples with questionable
intensity. Mark the duplicate sample.
•
• Sample Intensity Intensity scale
• codes (preliminary) final
• A 1 = very weak
• B 2 = weak
• C 3 =medium strong
• D 4 = strong
• E 5 = very strong
• F
• G
• H
• Final ranking
• Sample codes
•
• weakest strongest
• intensity intensity

• Duplicate:

289
 Aids for beginners

• Avoid too frequent re-tasting;

remember that the first impression is
usually correct.
• Rinsing the mouth thoroughly with
water at 35 'C may be a great help when
judging intensifies.
• Start rinsing with the first sample, do
not wait until fatigue sets in.

290
 Presentation of results at the blackboard

• During courses with 15 participants, the results

of five test subjects are written on the
blackboard.
• From these it can be seen whether or not the
results are the same or varied.
• Then a line is drawn and the correct ranking
order is presented by the panel leader.

291
 Examples of results at the blackboard

• Results B F D A H E G C
• of Participants H F D A B E C G
• B F D A H E C G
• B F D A H G E C
• B D F A C H E G
• Correct order B D F A H E C G

292
 Presentation of results at the notice board

• In this case, it is easier to survey

results when correct ones are
marked with a plus sign (+) and
incorrect ones with the given letter
(Table 11.2).

294
 (b) Ranking test with sodium chloride solutions

• Concentrations: 0.08, 0.09, 0.1, 0,1, 0.2, 0.3, 0.4,

0.5% (w/v).
• Sample preparation: see Table 9.2, p. 166.
• Example of results: see Table 11.2, p. 256.
• (c) Ranking test with citric acid solutions
• Concentrations: 0.003, 0.01, 0.02, 0.03, 0.03,
0.04, 0.05% (w/v).
• Sample preparation: see Table 9.3, p. 167.

295
 RANKING TEST IN ORDER OF DEGREE OF
PREFERENCE

• Samples suitable for this test are different

kinds of food.
• The simplest way is to buy competitive
products in supermarkets.
• Differently flavoured samples (such as
chocolate) can also be used as shown
below.

296
 • Number of samples
• Do not use more than three to four samples for
beginners.
• Also, in real tests, limit the number to four to
five to avoid fatigue.
Coding
• This is better done with letters than with
numbers. This is not true, so we will use
numbers of three digits.

297
 Test instructions

• The samples are tested in a specified

order. Those which are liked best are
placed to the left, those liked least to the
right.
• An example of a test sheet is shown in
Table 11. 11. The best liked sample gets
the first place in ranking, the least liked the
last (fourth).
• The other samples are ranked in between.

298
 Evaluation of the results

• This can be done as shown in Table 11.12.

The given rank positions (1 = first, 2 =
second, 3 = third, 4 = fourth) are listed below
the sample code.
• This procedure is, however, too time
consuming to be used during courses.
• The rank sum is easier and faster to
determine by using the tabulation chart of
Milhlhausen as shown in Table 11.13a.

299
 • The rank positions (1, 2, 3, 4) are
written in a column on the left side,
and the box head contains the
sample codes (916, 517, 480,650).
• The results of the participants are tallied.
• When the rank order 517, 480, 916, 650
has been given by a participant:
• the stroke for 517 is written in the box for
the first the stroke for 480 is written in
the box for the second

300
 • the stroke for 916 is written in the box for the third
the stroke for 650 is written in the box for the
fourth rank position (see Table 11.13a).
• After the results of all participants have
been entered in the various boxes of the
evaluation chart,
• the number of the results in each box are
added together and then multiplied by the
rank position number (first rank position
number:
X 1, second: X 2, third: X 3, fourth: X 4).

301
• The rank sum for each sample is obtained by
adding the single sums in each column (see
Table 11.13b).
• The lowest rank sum indicates the best liked
sample, the highest rank sum the least liked
sample.
• The rank sums of Tables 11.12 and 11.13b
are identical, but have been obtained in
different ways.

302
 Table 11.11-Ranking the degree of preference

• Test product: chocolate Name:

• Test subject No.: Date:
• You will receive four samples. Please rank them in order of your degree of preference.
• Write to the very left the code of the sample you like best (to which you give the first
position in the degree of preference) and to the very right the code of the sample you
like least.
• Mark the degree of preference for the other samples in between.
• ---------------------------------------------------------------------------------
• Sample Code
•
• ---------------------------------------------------------------------------------------------------------
• 1. rank 2. rank 3. rank 4. rank
• position position position position
• (most liked) (least liked)
• ----------------------------------------------------------------------------------------------------------

303
 Table 11.11-Ranking the degree of preference (Example for a
filled out test sheet)

• Test product: chocolate Name:

• Test subject No.: Date:
• You will receive four samples. Please rank them in order of your degree
of preference.
• Write to the very left the code of the sample you like best (to which you
give the first position in the degree of preference) and to the very
• right the code of the sample you like least.
• Mark the degree of preference for the other samples in between.
• ----------------------------------------------------------------------------
• Sample Code
• 517 480 916 650
• ------------------------------------------------------------------------------------------------
• 1. rank 2. rank 3. rank 4. rank
• position position position position
• (most liked) (least liked)
• ------------------------------------------------------------------------------------------------
•

304
• Table 11.12 - Results of ranking the degree of preference with
• differently flavoured chocolate samples
• --------------------------------------------------------------------------------------
• Test Rank position given for the samples
• Subject -------------------------------------------------------------------------
• No 916 517 480 650
• ---------------------------------------------------------------------------------------
• 1 3 1 2 4
• 2 1 3 2 4
• 3 2 1 3 4
4 4 2 1 3
• 5 1 2 3 4
• 6 2 1 3 4
• 17 3 1 4 2
• ---------------------------------------------------------------------------------
• Rank sums 44 21(**) 42 63(**)
• of the samples

305
 Statistical analysis of the data

• Table 11.14 shows whether or not the result is significant

• and Table 11.15 shows whether or not it is highly significant
•
Table11.13a-Tabulation chart for ranking tests (with listing of one result)
• ----------------------------------------------------------------------------------------
• Sample codes
• 916 517 480 650
• -----------------------------------------------------------------------------------------
• 1. Rank position /
• -------------------------------------------------------------------------------------------
• 2. Rank position /
• -------------------------------------------------------------------------------------------
• 3. Rank position /
• -------------------------------------------------------------------------------------------
• Rank position /

306
 Table 11.13b - Ranking the degree of preference with chocolate
Samples (Use of the tabulation chart of Table 11.13a)
Sample codes
• 916 517 480 650
• (a) (b) (a) (b) (a) (b) (a) (b)
• ---------------------------------------------------------------------------------
• 1. Rank // ///// ///// /
• position ////
• (2X 1) = 2 (14X 1) =14 (IX 1) =1
• ---------------------------------------------------------------------------------------
• 2. Rank ///// // ///// /
• position / ///
• (6X2) = 12 (2X2) = 4 (8X2) =16 (1X2) =2
• 3. Rank ///// / ///// ///
• position / //
• (6X3)=18 (1X3) =3 (7X3) =21 (3X3 )=9
• ------------------------------------------------------------------------------------------
• ------------------------------------------------------------------------------------
• 4. Rank /// / ///// /////
• position ///
• (3X4) =12 (1X4) =4 (13X4) =52
• -------------------------------------------------------------------------------------------
• Rank sum 44 21** 42 63**
• of the 916 517 480 650
• samples
• -------------------------------------------------------------------------------------------
• (a)= number of judgments.
• (b)= rank sum of the respective rank position

• ----------------------------------------------------------------------------------------

307
 Detail from Table 11.14

• Rank totals required for significance at the 5% level

• (α ≥ 0.05)
• No. of Number of treatments or samples
ranked
• Repli 2 3 4 5 6 7 8 9
• (17) 22-29 27-41 32-53 38-64 43-76 48-88 53-100 58-112
• 22-29 28-40 35-50 41-61 48-71 54-82 61-92 67-103
• ---------------------------------------------------------------------
• The four-figure blocks in Table 11.4 represent:
• lowest insignificant rank sum,
• any treatment - highest insignificant rank sum,
• any treatment; lowest insignificant rank sum,
• predetermined treatment - highest insignificant rank
insignificant rank sum, predetermined treatment.
308
• For our exercise (number of responses = 17,
number of ranked samples = 4, and any
treatment), we find in Table 11.14 "32-53".
• This means that 32 is the lowest insignificant
rank sum and 53 the highest insignificant rank
sum.
• In our exercise, sample 517 had the lowest rank
sum: 2l.
• This rank sum is not only lower than 32, from
Table 11.14, but also lower than 30, from Table
11.15. This means that the placement of sample
517 in the first rank position is "highly significant"

309
 • or, in other words; among the four
chocolate samples, sample 517 is liked
best.
• The placement of sample 650 in the
fourth rank position, with the rank sum of
63, is also highly significant.
• The highest rank sum, from Table 11.15, is
55. This means that among the four
chocolate samples, sample 650 is liked
least.

310
 Detail from Table 11.15
• Rank totals required for significance at the 1% level (α
≥0.01)
• No. of Number of treatments or samples ranked
• Replications ------------------------------------------------------------
2 3 4 5 6 7 8 9
• 17 20-31 25-43 30-55 35-67 39-80 44-92 49-104
53-117
• 21-30 26-42 32-53 38-64 43-76 49-87 55-98
60-110
• The rank sums of the samples 480(= 42) and 916 (= 44) are
within the ranges 32-53 (Table 11.14) and 30-55 (Table
11.15). Their rank position is not significant.

311
• If it is of interest, the ranking test can be
supplemented by a triangle test with these
two samples and the additional question of
preference asked.
• The results of ranking tests are not always
as clear as in our exercise.
• Variations may have several causes: the
samples were too similar in their character
(aroma. flavour);
• excessive demands were made upon the
test subjects (too many samples, fatigue
due to a lingering aftertaste); or none of the
samples were acceptable (liked).

312
313
 PREFERENCE TESTS

• Preference tests include the paired

comparison test, the multiple
comparison test, the hedonic scale,
and ranking tests.
• Sample questionnaires and examples
of these methods follow.

314
 Paired comparison test
• The paired comparison test used in
preference testing is similar to that used
for difference testing.
• When testing preferences, the panellist is
presented with two coded samples and is
asked which he prefers, such as in the
next questionnaire.

315
 QUESTIONNAIRE FOR PAIRED COMPARISON TEST

• NAME DATE
• PRODUCT
• Taste the two coded samples in the following order:
• 317 225
• Which of these two samples do you prefer?

• Comments:

316
 EXAPAPLE ON PAIRED PREFERENCE TEST

• To determine which of two spice combinations was

preferred in a drop cookie, a paired comparison
preference test was used. Fifty panellists compared
the two cookies.
• Twenty-five panellists tasted a cookie from treatment
A first and the other 25 tasted a cookie from
treatment B first.
• Thirty-five of the fifty panellists preferred the cookies
from treatment B.
• According to statistical Chart, this indicates a
significant preference at the 1 % level.
• If 33 persons had preferred one cookie over the other,
it would have been significantly preferred at the 5%
317
level.
 Multiple Comparison Difference Test
Questionare for difference test

318
319
320
 Ranking the degree of preference
• Test product: chocolate Name:
• Test subject No.: Date:
• You will receive four samples of chocolate having the code Nos:. 517,
480, 650, and 916.
• Please rank them in order of your degree of preference.
• Write to the very left the code of the sample you like best (to which you
give the first position in the degree of preference) and to the very right
the code of the sample you like least.
• Mark the degree of preference for the other samples in between.
• -------------------------------------------------------------------------------
• Sample Code
•
• --------------------------------------------------------------------------------------------
• 1. rank 2. rank 3. rank 4. rank
• position position position ( position
• (most liked) (least liked)
• ----------------------------------------------------------------------------------------------------------

321
 Hedonic Scale Test

• *The most commonly used scale for

preference testing is the nine-point
hedonic scale.
• The term "hedonic" is defined as "having
to do with pleasure."
• It should only be used in connection with
scales in which the panelist expresses his
degree of liking or disliking, such as in the
following sample questionnaire.
322
• QUESTIONNAIRE FOR HEDONIC SCALE

• NAME DATE

• PRODUCT

• Taste these samples and check how much you like or dislike each one

• 148 914 579 777

• like extremely like extremely like extremely like extremely
• like very much like very much like very much like very much
• like moderately like moderately like moderately like moderately
• like slightly like slightly like slightly like slightly
• neither like nor dislike neither like nor dislike neither like nor dislike neither like nor dislike
• dislike slightly dislike slightly dislike slightly dislike slightly
• dislike moderately dislike moderately dislike moderately dislike moderately
• dislike very much dislike very much dislike very much dislike very much
• dislike extremely dislike extremely dislike extremely dislike extremely
•
• Comments:

323
 EXAMPLE FOR HEDONIC SCALE TEST

• The ratings for each sample are given

numerical values ranging from like
extremely (9) to dislike extremely (1).
• The results are analysed by analysis of
variance, which has been already described
before.
• If only two treatments are evaluated, the
mean scores received by each can be
compared using the t-test.

324
 Analysis of variance

• Chewiness
• --------------------------------------------------------------------
• Brands
• --------------------------------------------------------------------
• Judges A B C D Total
• --------------------------------------------------------------------
• 1 2.2 1.9 3.2 0.2 7.5
• 2 2.5 1.0 4.2 2.0 9.7
• 3 3.3 3.0 4.0 1.5 11.8
• 4 2.6 5.4 4.2 1.3 13.5
• 5 2.6 4.3 5.1 0.1 12.1
• 6 2.2 3.4 4.7 1.9 12.2
• 7 2.4 3.6 5.5 3.2 14.7
• 8 1.6 3.9 5.7 1.8 13.0
• --------------------------------------------------------------------------
• Total 19.4 26.5 36.6 12.0 94.5
• ----------------------------------------------------------------------------
• Mean 2.4 3.3 4.6 1.5
325
 Correction factor (CF) = 94.52/32
= 8930.25/32
= 279.07
Sum of squares, samples=(19.42 + 26.52 + 36.62 + 12.02)/8 – CF
=2562.17/8 – CF
=320.27 - 279.07
=41.20
Sum of squares, judges =(7.52 + 9.72 + 11.82 + ... + 13.02)/4 – CF
=1152.17/4 – CF
=288.04 - 279.07
=8.97
Sum of squares, total = (2.22 + 2.52 + ... + 1 .82 – CF
=347.09 - 279.07
=68.02

326
• Analysis of variance table
• ----------------------------------------------------------------------------------
• So df SS MS F
• Variation
• ----------------------------------------------------------------------------------
• Samples3 41.20 13.73 16.1 5*
• Judges 7 8.971.281.51
• Error 21 17.85 0.85
• Total 31 68.02
• ---------------------------------------------------------------------------------

327
• The F value for samples is 16.15.
• According to Chart 3 in the Appendix, page 65, if
the F value exceeds 3.07 there is a significant
difference at the 5% level (*) and if it exceeds
4.87 there is a significant difference at the 1%
level (**).
• There is a significant difference at the 1% level
for samples so the value is marked with two
asterisks. There is no significant difference for
judges.

328
• Tukey's Test is used to determine which samples are signifi-
• cantly different. The standard error of the mean is calculated.
• SE = √0.85/8
• = √ 0.1062
• = 0.32
• The value from Chart 4 of the Appendix, page 67, for four
• samples and 21 degrees of freedom is 3.95.
• Least significant difference = 3.95 X 0.32 = 1.3
• Sample scores: A B C D
• 19. 26.5 36.6 12.0
• Sample means: 2.4 3.3 4.6 1.5

329
330
331
 DESCRIPTIVE SENSORY ANALYSIS

• A group of highly trained panellists

examine the flavour or texture of a
product to provide a detailed
descriptive evaluation of it.
• The most commonly known
descriptive methods are the flavour
profile and the texture profile.

332
 The flavour profile is the description of the
flavour and aroma of a food product.
 The description names the perceptible
factors,
 the intensity of each factor,
 the order in which the factors are perceived,
 aftertaste,
 and overall impression.

333
 • The texture profile is the description of
the textural characteristics perceived in a
food product, the intensity of each factor,
and the order in which they are perceived.
• Mechanical characteristics are
described qualitatively and quantitatively;

334
 • geometrical characteristics are
described qualitatively and semi-
quantitatively;
• the description of fat and moisture
characteristics depends on the product
being studied.
• Hardness, fracturability, chewiness
gumminess, adhesiveness, and
viscosity are mechanical properties.

335
 • Examples of geometrical properties are
grittiness, coarseness, and fibrousness.
• Descriptive analysis is a valuable tool in
difference testing and in product
development work.
• It provides a complete description of
sample differences and guides the
product developer in modifying product
characteristics to meet consumer
demands.

336
 • *The training of profile panels requires
considerable time and the members must
possess a high degree of motivation and
interest.
• Once trained, however, the panel can
provide thorough and reliable descriptions
of products in a short time.
• Since the descriptive panel members work
together as a group, forceful members
could have undue influence on the other
panellists and hence, change the results.

337
 • A method of sensory evaluation called
quantitative descriptive analysis was
developed at the Stanford Research Institute.
• During preliminary sessions the sensory
properties of the product are identified by a
trained panel.
• Samples are made up to illustrate the
different properties so that the panel agrees
on the meaning of each term used.

338
 • During these sessions the judges work
together as a group and discussion is
encouraged.
• *The panel members decide the sensory
properties that should be evaluated and they
also develop the language to be used.
During evaluation sessions the panellists
work individually.

339
 • *The scale used is an interval scale
consisting of a horizontal line (15 cm) long
with anchor points (1.3 cm) from each end
and usually, but not necessarily, having a
mid point.
• Each anchor point is usually labelled with a
word or expression.
• A separate line is used for each sensory
property to be evaluated.

340
 • Each judge records his evaluation by
making a vertical line across the horizontal
line at the point that best reflects his
perception of the magnitude of that
property.
• After the panelists have completed their
judgments, the experimenter superimposes
a grid dividing the line into 60 units, to
assign a number between 0 and 60 to each
rating. These values are then tabulated.

341
 QUESTIONNAIRE FOR DESCRIPTIVE ANALYSIS
WITH SCALING

• NAME DATE
• Please evaluate the firmness of these sample of frankfurters.
• Make vertical lines on the horizontal line to indicate your rating of
the firmness of each sample.
• Label each vertical line with the code number of the sample it
represents.
• Please taste the samples in the following order:
• 572 681 437 249

• ___________________________________

•
• very soft very firm
•

342
 Definition of texture and flavour

• Both of these terms are not only frequently

used in the literature, but are also defined in
different ways. This makes their
understanding difficult.
• 1. DEFINITION OF CONSISTENCY,
STRUCTURE AND TEXTURE
• These terms were defined in such a way that
they can be differentiated clearly. Physical
properties are limited by their connotation in
consistency, structure and texture.

343
 • These terms cannot be used synonymously
even though all can be characterized
rheometrically by similar force action
methods.
• * Rheology was defined as the science of
deformation and flow of substances.
• All substances flow only their flow
behaviour is different.
• After deformation, the particles return to
their original position.

344
 (a) Definition of consistency

• Consistency describes a substance's

compactness, the connection of its
elements, density, firmness or
viscosity, and its resistance or plastic
behaviour (action) during continuous
changes of form.
• Its plastic behaviour can be described
as a deformation curve, forming curve
or flow curve in the case of viscous
substances.
345
 (b) Definition of structure

• Structure describes a substance's set-up,

construction, granular make-up, textile
construction (as macrostructure,
microstructure and molecular structure) and,
depending on network structure or structure
power, as liquid, semi-firm and firm foods.
• The terms 'firmness' or 'softness' are also
used.
• Even though not exactly defined, they
descriptively evaluate consistency and
structure as well as density and hardness
properties of substances.

346
 • The crosslinks of polymers
considerably influence the firmness of
these high molecular weight
substances.
• In the case of products of plant origin
the fruits and bulbs (tubers) evaluated
as cell turgor of the cell condition.

347
 (c) Definition of texture

• *Texture describes a substance's tissue

structure. The term is derived from the
word textile. Textile manufacturers
produce, through spinning and weaving,
materials of various 'textures'.
• *Psychophysical investigations made by
food technologists about 30 years ago led
to a classification of hedonic sensations
perceived through the mouth which are
important to the modern consumer.

348
 • In modern sensory analysis, texture is
characterized by detailed subdivisions of the
help of tactile and pressure sensations
perceived through biting, chewing and
swallowing.
• This has led to a deeper understanding of
the texture complex of foods.

349
 • When judging foods by tasting, properties
such as softness in biting, ease of
disintegration into fragments, connection of
the fibres, and other pressure and softness
sensations on the tongue, hard palate and
cheeks are considered.
• An easy to bite, tender, firm but soft and
juicy texture seems to be the most
acceptable to the consumer. As can be
seen, sensual enjoyment is derived from
functional properties.

350
 • The following definition have been
proposed for texture:
• "Texture is the composite of those
properties which arise from the physical
structural elements and the manner in
which it registers with the physiological
senses".
• The term texture, therefore, encompasses
all properties of foods which are perceived
by the kinesthetic and tactile senses in the
mouth, e.g. tenderness, density, granular
structure, fragility, humidity, etc.
351
 Following are 96 texture descriptions and material
properties

352
 • *The importance of texture as a 'sensory quality
complex' is very well explained in the
introduction of the mentioned publication.
• *Texture and juiciness are of utmost, top-rate
importance for the palatability (eating value) of
meat and other foods.
• Prepared, ready-to-eat meat must be first of all
tender and juicy.
• Too firm, tough, coarse or dry meat texture is not
appealing and it will always be rated low in
consumer tests."

353
 • *Among all quality properties, the texture of
many foods is the most sensitive (delicate) and
most changeable, especially during cold storage
and freezing-thawing processes, and in all
cooking processes.
• In prepared meat, a high texture quality is related
to excellent culinary art and precise time-
temperature cooking.
• Careless routine preparation can change an
originally tender piece of meat into a dry, leather-
like product.

354
• *The 96 descriptive texture designations
and product properties listed can be
classified by adding intensifies to the
descriptions, e.g. 'little - slight - very -
strong - complete - predominant' etc. The
range of this descriptive scaling is,
however, variable and subjective.

355
• *It is most difficult to determine precisely
the border values of 'tender - almost
tender- slightly tender, etc.' while chewing.
In principle, each scaling point should
have the same meaning for each test
subject. This is only possible when work
is done in an objectified way.

356
 Analytical texture evaluation

• *For a sensory record of textural features

of any food, the different physiological,
tactile and pressure sensations are
analytically evaluated in sequence of their
perception, and their intensifies defined.
The bite, chew and swallow process is
divided into three stages.

357
• *Depending upon the product, 5 to 6 scale
points are used for the mechanical
perceptions in the mouth. Over 20
mechanical and kinaesthetic properties
are evaluated sensorial in sequence of
their perception and their intensifies
defined.

358
• *The following features (characteristics) belong
to the mechanical properties: hardness,
cohesiveness', adhesively, density and viscosity,
factorability, cheesiness and gumminess
(referring to hardness and connection of
particles).
• *Geometrical features are particle size, hardness
and fibre orientation.
•

359
• Texture profile analysis
• *The mentioned analytical texture evaluation has
been worked out in the research laboratories of
the General Foods Corporation, and is known as
General Foods Texture Profile Analysis (TPA). It
is used in research and development in
American, Japanese and Latin American
countries. Sensory reference standards form the
basic for the scale points.

360
• *There exist, for example, nine steps for
hardness, seven steps for brittleness, seven
steps for cheesiness, etc. Panel training and
selection is explained and described in detail.
The texture analysis belongs to the so-called
'high school' of sensory analysis. The TPA
requires knowledge of the product and of the
physiology of chewing, as chew number, chew
speed and amount of saliva are related to
texture.
361
• *There exist instrumental methods for
evaluating texture, and over one hundred
instruments, such as the shear apparatus,
compress meters, electrometers, cuterr
devices, consist meters and
viscosimeters which measure various
physical properties, but which are not
always correlated with sensory results.

362
• *For measuring the various texture profile
features (characteristics),
• the classical General Foods Texturometer
is used.
• Exact measurements are made according
to a compression principle (squeezing and
comminuting) providing texture
profilograms.
363
• *Sensory and instrumental texture
meacerement is a special psycho
• physical (and auditive) science and is
open to all those with real interest. It
requires, however, profound knowledge of
the specific product and sensitivity.

364
 • 2.FLAVOUR
• Definition
• *Two definitions can be found in literature. In the first case, the
• term flavour includes the four basic tastes and the aroma
• perceived through tasting.
* Standards organizations of different countries increasingly
• use the second definition:FLAVOUR- total impression of taste,
• odour, tactile, kinesthetic, temperature and pain sensations
• perceived through tasting.

365
 *The cooperation of the various factors in aroma and
flavour development is illustrated very well in the following Fig.

366