SOP: Hedonic Rating Test

GIFSQ — Global Institute for Food Safety & Quality

Principle

Hedonic rating measures pleasurable or unpleasurable experiences — the consumer acceptability of food products. From one to four samples are served to the panelist at one session (a larger number, 5–10, only if the samples are mild-flavoured or are rated for colour or texture only). The panelist rates the acceptability of the product on a scale, usually of 9 points, ranging from “like extremely” to “dislike extremely”; scales of different ranges or other experience phrases may also be used. The results are analysed for preference using data from large untrained panels. Semi-trained panels in smaller numbers are used to screen a number of products for selecting a few for consumer preference studies. The test is more useful for determining preferences than for determining differences — two samples that differ clearly in flavour may still both receive “neither like nor dislike” from the same panelist.

Apparatus required

• Individual judging booths (identical, uniformly illuminated), each fitted with drinking water, a glass, clean towels and a basin or receptacle for expectoration
• Coloured lighting (where needed to mask small colour differences between samples)
• Stainless steel, glass or china serving dishes and cups; plain serving trays; small stainless steel forks and spoons
• Prepared evaluation cards for the hedonic rating test (one card per product when pronounced after-effects preclude a second sample or when independent judgements are sought; multi-column cards when relative preference is studied)

Sample preparation

Prepare all samples in exactly the same manner. Code the samples with three- or five-digit numbers. When pronounced after-effects are met with, precluding the testing of a second sample, or when independent judgements are sought, use a separate card for each product; when relative preference is the object, use cards with multiple columns for the number of test samples.

Procedure

Serve the coded samples to the panelist (semi-trained panel: 10–25 judges; 5–10 samples per session; a larger number only if the samples are mild-flavoured or are rated for colour or texture only)

↓

Ask the panelist to taste each sample and check how much it is liked or disliked

↓

For each sample, the panelist checks the point on the scale that best describes the feeling about the sample: Like extremely / Like very much / Like moderately / Like slightly / Neither like nor dislike / Dislike slightly / Dislike moderately / Dislike very much / Dislike extremely

↓

Ask the panelist to give a reason for this attitude

↓

Collect the signed cards

Interpretation of results

Assign numerical values to each point on the scale — 1 point usually for “like extremely” and 9 points for “dislike extremely”, or vice versa (the scale may be reduced to 7 or 5 points). Average the scores received by each sample and compare with the average score of the other samples in the series. The hedonic ratings may also be evaluated by the ranking method: arrange the ratings given by each panelist to the samples in increasing or decreasing numerical order, assign ranks (averaging tied ranks), total the rank sums per sample, and refer to the rank-sum tables (Tables 19-5a and 19-5b). A more accurate evaluation is the fiducial-limits method: the fiducial limits represent the range of average scores (above and below the average score of the control) within which an average score for a test sample is not significantly different from the control; scores above or below the limits are significantly different.

Fiducial limits procedure:

1. Determine the standard error of the mean of the reference sample
2. Calculate the limits L = SE × t, where t is found from Table 19-11 for the degrees of freedom at the chosen significance level
3. Determine the fiducial limits FL = mean score of the control ± L

4. Classify the test samples as superior, inferior or not significantly different from the control according as their hedonic mean scores fall above, below or within the fiducial limits

5. Note: the standard error expression in the available copy of the book reads SE = s/√(n − 1) (standard deviation divided by the square root of n − 1); carried as printed.

Worked illustration from the book (ranking evaluation): a factory sample of orange marmalade was rated on the 9-point hedonic scale against three competitive brands by 10 panelists. Mean scores: A 2.3, factory 2.8, B 5.1, C 5.6 — the quality order is A > factory > B > C. Converting to ranks gave rank sums of 17 (factory), 13 (A), 33 (B) and 40 (C). For 4 treatments and 10 replications, Tables 19-5a/b give: at the 5% level a rank sum of 16 or lower, or 34 or higher, is required for significance; at the 1% level, 14 or lower, or 36 or higher. Sample A (13) is significantly superior at the 1% level; sample B (33) is significantly inferior at the 5% level; sample C (40) is significantly inferior at the 1% level; the factory sample is qualitatively placed between sample A and samples B and C. Re-ranking the factory sample against sample A alone (rank sums 17 and 13) showed them comparable in quality.

Worked illustration from the book (fiducial limits): for the factory (reference) sample, Σx = 28, Σx² = 82, n = 10; mean score x̄ = 28/10 = 2.8; variance s² = 0.36; standard deviation s = 0.6; standard error of the mean SE = 0.6/√(10 − 1) = 0.2. t (Table 19-11, 9 degrees of freedom) = 2.262 at the 5% level and 3.250 at the 1% level. Limits L = SE × t = 0.2 × 2.262 = 0.4524 at 5% and 0.2 × 3.250 = 0.65 at 1%. Fiducial limits = 2.8 ± 0.45 = 2.35 to 3.25 at 5%; 2.8 ± 0.65 = 2.15 to 3.45 at 1%. Conclusion: sample A is significantly superior, and samples B and C are significantly inferior, to the factory sample at the 1% level. The fiducial-limits method may also be used to evaluate paired comparison results.

Specimen evaluation card fields: code (per sample column); the 9-point scale with check-points; reason for the attitude; signature. The card reminds the panelist: “Remember you are the only one who can tell what you like. An honest expression of your personal feeling will help us.”

Precautions

• Use separate cards for each product when pronounced after-effects preclude a second sample or when independent judgements are sought.
• Remember that the test measures preference, not the size of differences — do not use it to establish that two samples differ.
• Conduct judging in individual booths so that judgements are independent; allow no communication between panel members except consultation with the panel leader on any point of doubt.
• Test in the hour after a normal meal (e.g., 10.00 a.m. to 12 noon, or 3 to 5 p.m.).
• Panel members should not smoke or chew pan or supari for at least half an hour before the test.
• Judge quickly but not hurriedly; make odour observations by sniffing before tasting.
• Allow retesting, but avoid prolonged testing; tasters may swallow or spit out samples, and should rinse between samples (or use bread or puffed rice) to remove a strong flavour.
• Keep the evaluation card simple, clearly typed or printed, with unambiguous terms and directions in the desired sequence of action.

References

Ranganna, S., Handbook of Analysis and Quality Control for Fruit and Vegetable Products, 2nd edition.

An Introduction to Taste Testing of Foods, Merck Technical Bulletin, Merck Chemical Division — cited in the book.