Sensory Evaluation: Types of Tests

This lesson covers the different types of sensory evaluation tests used in food development and assessment, as required by the AQA GCSE Food Preparation and Nutrition specification (8585), section 3.5. You need to understand the purpose, method and application of preference tests, discrimination tests and grading tests.

---

What Is Sensory Evaluation?

Sensory evaluation (also called sensory analysis or sensory testing) is the scientific method of using human senses to evaluate food products. It involves systematically assessing the:

• Appearance — colour, shape, size, gloss
• Aroma/Smell — the smell of the food before and during eating
• Taste — the five basic tastes (sweet, sour, salty, bitter, umami)
• Flavour — the combination of taste and aroma
• Texture/Mouthfeel — crunchiness, smoothness, chewiness, crumbliness
• Sound — crunch, sizzle, snap

Why Is Sensory Evaluation Important?

Sensory evaluation is used by:

User	Purpose
Food manufacturers	Developing new products; improving existing products; quality control; comparing with competitors
GCSE students	Evaluating practical dishes; developing recipes; the NEA (Non-Exam Assessment) tasks
Researchers	Investigating flavour interactions; consumer preferences; the effect of ingredient changes

---

Three Categories of Sensory Tests

Sensory tests are grouped into three main categories:

flowchart TD
    A["Sensory<br/>Evaluation Tests"] --> B["PREFERENCE Tests<br/>Which do you prefer?"]
    A --> C["DISCRIMINATION Tests<br/>Can you tell the<br/>difference?"]
    A --> D["GRADING Tests<br/>How much do you<br/>like it? / Rank them"]

    B --> B1["Paired Preference"]
    B --> B2["Hedonic Scale"]

    C --> C1["Triangle Test"]

    D --> D1["Ranking"]
    D --> D2["Rating"]
    D --> D3["Sensory Profiling<br/>(Star Diagram)"]

    style A fill:#2c3e50,color:#fff
    style B fill:#3498db,color:#fff
    style C fill:#e74c3c,color:#fff
    style D fill:#27ae60,color:#fff

---

1. Preference Tests

Preference tests determine which product a taster prefers. They are subjective — based on personal opinion.

Paired Preference Test

Feature	Detail
Purpose	To find out which of two samples a taster prefers
Method	Taster is given two coded samples and asked "Which do you prefer?"
Number of samples	Two
Result	Simple — either Sample A or Sample B is preferred
Advantage	Quick and simple to carry out; easy to analyse results
Limitation	Only compares two products; does not tell you why one is preferred or how much it is preferred
Example use	Comparing an original recipe with a modified version (e.g., reduced-sugar biscuit vs original)

Hedonic Scale Test

Feature	Detail
Purpose	To measure how much a taster likes or dislikes a product
Method	Taster rates a sample on a scale from "dislike extremely" to "like extremely" (usually 5, 7 or 9 points)
Scale example	1 = Dislike extremely, 2 = Dislike, 3 = Neither like nor dislike, 4 = Like, 5 = Like extremely
Result	Numerical data that can be averaged and compared
Advantage	Shows the degree of liking; can compare multiple products; produces quantitative data
Limitation	Subjective — different people interpret the scale differently; may be influenced by mood
Example use	Testing consumer reaction to a new product; comparing several flavours of yoghurt

Example of a 5-point hedonic scale:

Score	Description
5	Like extremely
4	Like slightly
3	Neither like nor dislike
2	Dislike slightly
1	Dislike extremely

Exam Tip: The hedonic scale produces numerical data that can be averaged and compared. This is a key advantage over the paired preference test, which only gives a binary (A or B) result.

---

2. Discrimination Tests

Discrimination tests determine whether a taster can detect a difference between samples. They are more objective than preference tests.

Triangle Test

Feature	Detail
Purpose	To determine whether a taster can detect a difference between two products
Method	Taster is given three coded samples — two are the same, one is different. The taster must identify the odd one out
Number of samples	Three (two identical, one different)
Result	The taster either correctly or incorrectly identifies the odd one out
Probability of guessing correctly	1 in 3 (33%) — so results must be significantly above this to prove a real difference
Advantage	Objective; determines if a difference can be detected; useful for quality control
Limitation	Does not indicate preference or the nature of the difference; taster fatigue with strong flavours
Example use	Testing whether consumers can tell the difference between a product made with butter vs margarine; checking batch consistency

flowchart LR
    A["Three Samples<br/>Presented"] --> B["Sample 247<br/>(Product A)"]
    A --> C["Sample 391<br/>(Product B)"]
    A --> D["Sample 582<br/>(Product A)"]

    B --> E["Taster identifies<br/>the ODD ONE OUT"]
    C --> E
    D --> E

    E --> F{"Correct?"}
    F -- "Yes" --> G["Difference<br/>detected"]
    F -- "No" --> H["No difference<br/>detected"]

    style C fill:#e74c3c,color:#fff
    style G fill:#27ae60,color:#fff

How the Triangle Test Is Set Up