Measuring emotions in commercial products

May Ng

Emotional measures GO beyond liking

If liking is the first and only thing that came to your mind?

Brief Intense Often focused on a referent

WHATis emotion? King & Meiselman

(2010)

Example: The chocolate made her happy

Activated Pleasant 

Pleasant   Unpleasant 

Unactivated Unpleasant 

Unactivated Pleasant 

Low Activation

High ActivationActivated Unpleasant 

450450

Larsen & Diener (1992)

Engagement

EMOTIONMODEL

Multidimensional

Liking data is not enough

Emotional quality of products is important for: – Differential advantage – Purchase decision

WHY measure

emotion?

Examples: EsSense Profile Consumer defined

Check-All-That-Apply (CATA)

HOWto measure

emotion?

EsSense Profile (King & Meiselman., 2010)

Overall acceptability (9 point scale)

Emotion terms (5 point scale)– 39 terms; most positive – Selected from published psychological emotion list

Reference:King, S. C., & Meiselman, H. L. (2010). Development of a method to measure consumer emotions associated with foods. Food Quality and Preference, 21 (2), 168-177

OBJECTIVES

EsSense Profile

Consumer defined CATA

COMPARE v.s.

EMOTION data GO BEYOND LIKING???

Samples

Subjects

Methodologies

11 UK Commercial Blackcurrant Squashes

Labeled as P1- 113

No Added Sugar

2 NicheAdded Sugar

6Added Sugar

Group 1n=100

METHODS &MATERIALS

Group 2n=100

EsSense Profile

Consumer defined CATA

Methodologies

EsSense Profile Consumer defined CATAPredetermined psychological

list

Consumer self generated & defined

list (1:1 Rep Grid Interview; n=29)5 point scale

(Quantitative data) Check-All-That-Apply(Qualitative data)

9 – point scale to measure liking

METHODS &MATERIALS

EsSense Profile (39) C-defined CATA (36) 25

POSITIVE(+ve)

AdventurousActive AffectionateCalmEnergetic EnthusiasticFreeFriendlyGladGoodGood-naturedHappyInterested Joyful Loving Merry Nostalgic Peaceful Pleasant

3 NEGATIVE

(-ve) Bored Disgusted Worried

11 UNCLEAR

(+/-ve)Aggressive Daring Eager Guilty Mild PoliteSteady Tame Understanding Wild

16 POSITIVE

(+ve)Approval At ease Attentive Comforted Curious Desire GoodHappyInterested Pleasant Pleased Refreshed Reminiscence SatisfactionTrust Warm

19NEGATIVE

(-ve)Angry Annoyed Bored Cautious Confused DisappointmentDiscontented Disgusted DispleasureNot refreshed Regret Resentment Sceptical Shocked Sickly Uncomfortable UnhappyUnpleasant Worried

1 UNCLEAR

(+/-ve)Guilty pleasure

PleasedSatisfied SecureTender Warm Whole

EMOTION LEXICONS

25POSITIVE

(+ve)AdventurousActive AffectionateCalmEnergetic EnthusiasticFreeFriendlyGlad

GoodGood-natured

HappyInterested Joyful Loving Merry

Nostalgic Peaceful

Pleasant

3 NEGATIVE

(-ve)

Bored DisgustWorried

11 UNCLEAR

(-/+ve)Aggressive Daring Eager

Guilty Mild PoliteSteady Tame Understanding Wild

16 POSITIVE

(+ve) Approval At ease Attentive Comforted Curious Desire

GoodHappyInterested Pleasant Pleased Refreshed

Reminiscence SatisfiedTrust

Warm

19 NEGATIVE

(-ve) Angry Annoyed

Bored Cautious Confused DisappointmentDiscontented

Disgust DispleasureNot refreshed Regret Resentment Sceptical Shocked Sickly Uncomfortable UnhappyUnpleasant

Worried

1 UNCLEAR

(+/-ve)

Guilty pleasure

PleasedSatisfiedSecureTender

Warm Whole

Reminiscence

EsSense Profile (39) C-defined CATA (36)

EMOTION LEXICONS

DATA ANALYSIS

Quantitativedata

Qualitativedata

EsSense Profile C-defined CATA

ANOVA and Tukeys Multiple Comparison test

to determine which products differed from each other

Principal Component Analysis (PCA)

Emotion mean datato determine product positioning

Consumer liking as supplementary variable.

Chi square testto determine which products differed from

each other

Multiple Correspondence Analysis

(MCA)Individual responses to CATA question

Products as supplementary variableto determine product positioning

ANOVA &

Tukeys Multiple Comparison liking scores

Results:EsSense Profile

RESULTS: EsSense Profile Liking data

11 UK commercial blackcurrant juice

squashes (Labelled as P1 – P11)

No Added Sugar

Niche Added Sugar

Added Sugar

Product Liking scores

Group

4 4.1 A3 4.3 AB10 5.0 BC8 5.3 CD6 5.5 CDE5 6.0 DEF2 6.0 DEF1 6.3 EFG9 6.5 FG7 6.7 FG11 6.8 G

Incr

ease

d in

Lik

ingSimilar

low liking scores

Similar high liking

scores

Do products with similar liking scores induce different emotional responses??

ActiveAdventurousAffectionate

Bored Calm

Daring

Disgusted

Eager

Energetic

EnthusiasticFree

FriendlyGlad

Good

Good‐ natured

Happy

InterestedJoyful

Loving

Merry

Peaceful

Pleased

PleasantPoliteSatisfiedSecure

Steady

Tame

Tender

UnderstandingWarm

Whole

Worried

Liking

‐1

‐0.75

‐0.5

‐0.25

0

0.25

0.5

0.75

1

‐1 ‐0.75 ‐0.5 ‐0.25 0 0.25 0.5 0.75 1

F2 (5

.58 %)

F1 (83.70 %)

Variables (axes F1 and F2: 89.28 %)Correlation circle:29/ 39 emotion terms were found significant in blind assessment

22 +ve & 3+/- ve3 –ve

‘Tam

e’

RESULTS: EsSense Profile

PCA:

1

2

3

4

56 7

8

9

10

11

‐5

‐80 ‐64 ‐48 ‐32 ‐16 0 16 32 48 64

Biplot (axes F1 and F2: 89.28 %)

Liking

NEGATIVE POSITIVEHigh liking Low liking

RESULTS: EsSense Profile

Low liking scores:

2

4More ‘Tame’

High liking scores:

No Added Sugar

Niche Added Sugar

Added Sugar

15

Less 'Adventurous’ & Less ‘Daring’

Q: Do products with similar liking scores induce different emotional responses?

A: Yes. There is a significant difference for some emotions.

11975

10

2

43

The difference in mean data is ≤ 0.5 point over a 5 point scale.

RESULTS: EsSense Profile

Results:C-Defined CATA

Product Liking scores

Group

3 4.06 A10 4.26 A4 4.62 AB8 4.81 AB6 4.83 AB2 5.45 BC5 6.0 CD9 6.36 D11 6.57 D1 6.61 D7 6.62 D

RESULTS: C-Defined CATALiking data

11 UK commercial blackcurrant juice

squashes (Labelled as P1 – P11)

No Added Sugar

Niche Added Sugar

Added Sugar

Similar low liking

scores

Similar high liking

scores

Incr

ease

d in

Lik

ing

Do products with similar liking scores inducedifferent emotional responses??

No Added Sugar

NicheAdded Sugar

Added Sugar

PLEASANT UNPLEASANT

LOW

AC

TIVA

TIO

NH

IGH

High liking Low liking

MCA:

RESULTS: C-Defined CATA

Activated Pleasant 

Pleasant   Unpleasant 

Unactivated Unpleasant 

Unactivated Pleasant 

Low Activation

High ActivationActivated Unpleasant 

450450

Larsen & Diener (1992)

Engagement

EMOTIONMODEL

Multidimensional

No Added Sugar

NicheAdded Sugar

Added Sugar

Higher activated pleasant emotion

Interested Discontent

Higher activated

unpleasant emotion

MCA:

Desire

Angry

RESULTS: C-Defined CATA

Low liking scores:

High liking scores:

No Added Sugar

NicheAdded Sugar

Added Sugar

1

Q: Do products with similar liking scores induce different emotional responses?

A: Yes. Key examples:

11975

1043 More ‘Attentive’

34More

‘Refreshed’

5More

‘Bored’

9Less ‘Reminiscence’Less ‘Warm’

Frequency count for ‘Refreshed’

33 consumers 18 consumers

43

Furthermore, consumers with same liking have different emotional responses .

RESULTS: C-Defined CATA

More consumers have selected

pleasant/ positive emotions

Same liking scores of 7 but

Different emotions

MCA: Product 7 (Highest Liking)

RESULTS: C-Defined CATA

ID16

ID33

Key Findings

EsSense Profile C-Defined CATA

Liking data is not enough

Emotions can be explained using degree of pleasantness and activation as illustrated by Larsen and Diener (1992)

KEY FINDINGS

Similar product/ emotion plot

EsSense Profile C-Defined CATA

Easy and straightforward to use

Better discrimination

Reported more negative emotions

Time consuming and limited statistical analysis

Limited number of negative emotion terms

PROS AND CONS

OTHER WORK...

Condition 1:Product

Condition 2:Packaging

Condition 3:Informedtasting

Can sensory properties evoke emotions?? YES!! Emotion data were also collected under different

conditions:

Take home message

Products with similar liking scores can induce different emotional responses

Exploiting emotional responses

is crucial to connect people to brandvia sensory experience

REMINDER: POSTER 

Measuring emotional response to food products

Eaton C1, Bealin-Kelly F2 & Hort J11Sensory Science Centre, University of Nottingham, UK.2SABMiller, Woking, UK.

Poster Board 1, Training room 1.

ACKNOWLEDGEMENTSupervisors: Dr Joanne Hort (UoN)Dr Carolina Chaya (UPM)

Acknowledge: Dr Ben Lawlor (GSK)Dr Tracey Hollowood(Sensory Dimensions)Prof Hal MacFie (Consultant) & all the volunteers who have taken part in my study!

Sponsor:

SPECIAL THANKS TO:The PFSG 2011 Student Travel Award

Thank you for your attentionAny Questions?

ContactsMay Ng : stxmln@nottingham.ac.uk Dr. Joanne Hort : joanne.hort@nottingham.ac.uk

• Reference:• Larsen, R. J., & Diener, E. (1992). Promises and problems with the ircumplex

model of emotion. Review of Personality and Social Psychology, 13, 25-59.

• Reference:• King, S. C., & Meiselman, H. L. (2010). Development of a method to measure

consumer emotions associated with foods. Food Quality and Preference, 21 (2), 168-177

Multiple Correspondence Analysis Popular technique to explore the relationships among

multiple categorical variable Data input: Individual CATA questions and 11 products

Principal Component Analysis

Methods & Materials

C-defined CATA Process of terms generation & selection Stage1: Terms elicitation

- 1:1 Rep Grid Interview (n=29)- Subjects presented with triads of products

289 365 972

Please indicate how two products differ in the same way from the third

Refreshing, premium,

guilty pleasure etc

etc

Stage 2: Defining constructs-Every subject has his/ her own list of constructs -Product assessment using individual constructs

109 Selection of terms based on word frequency table