Sensory Evaluation

C.S.Purushothaman Managing Director CREST PACKAGING PRIVATE LIMITED CONTRACT PACKAGING PROFESSIONALS

lecture delivered at the GRADUATE SCHOOL OF PACKAGING POST GRADUATE COURSE IN PACKAGING TECHNOLOGY CREST TRG AIDS

ACADEMIC YEAR 2001- 2002 Dec2001

1

SENSORY EVALUATION a scientific discipline used to measure, analyze and interpret reactions to those characteristics of materials as they are perceived by the senses of sight, smell, taste, touch and hearing.

CREST TRG AIDS

Dec2001

2

WHAT IS SENSORY EVALUATION? • A scientific discipline used

– evoke – measure – analyze – interpret • Reactions to characteristics of foods (and materials) as they are perceived by the senses

CREST TRG AIDS

Dec2001

3

SENSORY EVALUATION is the best available tool for
PRODUCT DEVELOPMENT
PRODUCT IMPROVEMENT
COMPATIBILITY STUDIES &
SHELF LIFE STUDIES

CREST TRG AIDS

Dec2001

4

SENSORY EVALUATION USES HUMAN BEINGS AS INSTRUMENTS HENCE IT IS GOVERNED BY PSYCHOLOGICAL FACTORS IN THE PROCESS OF ASSESSING

CREST TRG AIDS

Dec2001

5

HUMAN BEING AS INSTRUMENTS PANELISTS ARE : ❏Variable over time ❏Variable among themselves ❏Prone to Bias ❏Prone to fatigue and attention drift

CREST TRG AIDS

Dec2001

6

HUMAN BEING AS INSTRUMENTS TO MINIMISE THESE PITFALLS : ❏Measurements must be repeated ❏Number of panelists must be high ❏Biases must be avoided or reduced

CREST TRG AIDS

Dec2001

7

SENSORY EVALUATION USES HUMAN BEINGS AS INSTRUMENTS HENCE IT IS GOVERNED BY PSYCHOLOGICAL FACTORS IN THE PROCESS OF ASSESSING

CREST TRG AIDS

Dec2001

8

TYPES OF SENSORY TESTS
DISCRIMINATION TESTS


DESCRIPTIVE DESCRIPTIVE TESTS
AFFECTIVE TESTS

CREST TRG AIDS

Dec2001

9

DISCRIMINATION TEST IT DETERMINES HOW PRODUCTS ARE PERCEIVED AND THE SIGNIFICANCE OF DIFFERENCE

CREST TRG AIDS

Dec2001

10

DISCRIMINATION TEST
DETECTION OF SPECIFIC DIFFERENCE
SUITABLE FOR UNTRAINED OR TRAINED PANELISTS
PANELIST ONLY IF NO DISLIKE FOR PRODUCT
USED ONLY WHEN CHANGE IN FEATURE IS OBVIOUS
NO INFORMATION ON TYPE OF DIFFERENCE
REQUIRES 30 TO 50 PANELISTS
TOO HIGH & LOW NUMBER OF PANELISTS CAN BIAS
MORE THAN 100 PANELISTS CAN SKEW DATA

CREST TRG AIDS

Dec2001

11

DISCRIMINATION TEST
TRIANGULAR TEST
DUO TRIO TEST
PAIRED COMPARISON TEST
RANKING TEST CREST TRG AIDS

Dec2001

12

TRIANGULAR TEST B A

CREST TRG AIDS

A B A

A

•Three samples given •Two are same •One different •Identify ODD one

Dec2001

13

DUO TRIO TEST R

:

RB

Three samples given One labelled “R” ( reference) Other two coded One of the coded is identical to “R” Identify CORRECT sample Chance is 50 %

CREST TRG AIDS

Dec2001

14

PAIRED COMPARISON TEST AB •Pair of coded sample is given for some characteristic (sweetness, saltiness etc) •Identify the inferior sample •FEWER SAMPLES REQUIRED •Does not tell when both samples are bad

CREST TRG AIDS

Dec2001

15

RANKING TEST •Extension of Paired Comparison Test •Three or more coded samples are given •Rank for intensity of some characteristic

CREST TRG AIDS

Dec2001

16

DESCRIPTIVE TEST IT DETERMINES THE SPECIFIC DIFFERENCES AND HOW EXTREME ARE THOSE DIFFERENCES

CREST TRG AIDS

Dec2001

17

DESCRIPTIVE TEST
SUITABLE FOR TRAINED PANELISTS
TERMINOLOGY USED S COMMON & KNOWN TO ALL
EXPENSIVE AS TRAINING IS NEEDED
TIME CONSUMING
TRAINING IN PRODUCT NEEDED
PANEL SIZE AROUND 10
DETAILED INFORMATION CAN BE COLLECTED
QDA (Quality Descriptive Analysis)
FPA (Flavour Profile Analysis) CREST TRG AIDS

Dec2001

18

DESCRIPTIVE TEST
Linear Score Card

CREST TRG AIDS

Dec2001

19

LINEAR SCORE CARD •Many types of scoring systems are used •Can be unstructured, structured, or graphical •Scale of 7 to 10 points taken •Lesser than 7 not preferred

CREST TRG AIDS

Dec2001

20

QUANTIFYING THE ATTRIBUTES LINE SCALES Structured Scale

Unstructured Scale

Anchored or unanchored

CREST TRG AIDS

Dec2001

21

UNSTRUCTURED SCORE CARD

SAMPLE 1

2

3

4

5

6

SCORE

CREST TRG AIDS

Dec2001

22

STRUCTURED SCORE CARD SAMPLE No. ___ ___________9Highest Score ___________8 ___________7 ___________6 ___________5 ___________4 ___________3 ___________2 ___________1 Lowest Score

CREST TRG AIDS

Dec2001

23

GRAPHICAL SCORE CARD SAMPLE No. ___

CREST TRG AIDS

Dec2001

24

AFFECTIVE TEST IT MEASURES PREFERENCES FOR PRODUCTS OR MAGNITUDE OF LIKE / DISLIKE FOR A PRODUCT. CAN BE USED FOR CONSUMERS OR TRAINED PANELISTS

CREST TRG AIDS

Dec2001

25

AFFECTIVE TEST
Hedonic Test
Paired Preference Test (Multiple Comparison Preference Test)

CREST TRG AIDS

Dec2001

26

HEDONIC TEST •GREEK TERMINOLOGY •REFERS TO DEGREE OF LIKE OR DISLIKE •9 POINT SCALE •CAN BE COVERTED TO NUMERICAL

CREST TRG AIDS

Dec2001

27

HEDONIC SCORE CARD SAMPLE No. ___ ___________Like extremely ___________Like very much ___________Like Moderatly ___________Like Slightly ___________Neither like nor dislike ___________Dislike slightly ___________Dislike Mderately ___________Dislike very much ___________Dislike Extremely

CREST TRG AIDS

Dec2001

28

PAIRED PREFERENCE TEST Multiple Comparison Preference Test •SIMPLE & OLDEST •SAMPLES GIVEN •CONTROL SAMPLE WITH HEDONIC VALUE GIVEN •OFFERS A REFERENCE POINT AS A BASE •INDICATE PREFERENCE & WHY ?

CREST TRG AIDS

Dec2001

29

RADAR DIAGRAM for DIFFERENT BHELPURI SAMPLES

Sweet Paper

10

Sour

5

Green

Chilly Hot A

0 Stale

Rancid Raw

B C

Burnt Bitter

CREST TRG AIDS

Dec2001

30

QUANTIFYING THE ATTRIBUTES CATEGORY SCALES NUMERICAL 0 2.5 5 7.5 10 12.5 15

WORD None Very Slight Slight Slight- Moderate Moderate Moderate- Strong Strong

CREST TRG AIDS

Dec2001

HEDONIC Dislike extremely ----Neither Like / Dislike --Like Extremely

31

PANELIST MAKING OBSERVATION

CREST TRG AIDS

Dec2001

32

PANELISTS AT WORK

CREST TRG AIDS

Dec2001

33

PANELISTS AT WORK

CREST TRG AIDS

Dec2001

34

PRESENT DAY TREND

ELECTRONIC NOSE

CREST TRG AIDS

Dec2001

35