Consumer Product Rating
Sensory profiling in
conjunction with consumer rating data can be modelled to yield the relationship
between the desirability (“Overall liking”) of a product and the sensory
attributes. This reveals the importance of each attribute in enhancing the
desirability of the product; information that product designers can use to
improve products and make trade-off decisions.
In a quantitative research framework, target
consumers rate products on a number of different attributes, including overall
opinion. A typical question to assess taste for instance, would be worded as
follows:
How much do you like the taste of this product?
- I like it very much.
- I like it.
- I neither like it nor dislike it.
- I dislike it.
- I dislike it very much.
Products should be tested blind (unbranded).
Ideally, monadic in-use testing is recommended. Monadic, which
means only one product is tested by each respondent, is relatively expensive,
particularly if many products are to be tested.
Monadic testing is recommended because it eliminates
interaction effects and biases, and results can be compared with previous monadic
test results.
An alternative approach, Sequential Monadic,
may be adopted where respondents try one product and rate it, move to another product
and rate it, and then compare the two. Here the order sequence is usually randomized.
This approach is recommended when a forced comparison is required.
In this case the rating of the product that was tried first is pure monadic and provides a
measure for the initial response to the product.
Comparative testing, where a number of products are tried
alongside by respondents, is also used when respondents are required to compare products.
Comparative testing differs from sequential monadic in that all
samples in the assessment set are evaluated prior to giving a response.
Proto-Monadic testing is a variation of the monadic sequential
approach. In this scheme, two products are tested by each person, one at a time, with a monadic
measurement only on the product tried first. After the second product is tried, preference is measured,
both on an overall basis and for specific attributes.
This approach is well suited for testing product changes and for competitive
testing as it provides a strict single product measurement (monadic) as well as preference ratings.
Triangle Tests
Triangle tests are used
where the intent is to verify whether a change in composition can be detected.
This is required in case of cost reduction or a change of ingredients exercise.
For instance, a soft drink manufacturer may be interested to know whether a
reduction in the level of carbonation is noticeable or not.
In triangle tests, participants are asked to
identify which two of three products (‘item A’, ‘item A’ and ‘item B’) are the
same and which one is different. If the number of participants correctly
identifying the different item is statistically significant, it is concluded
that difference between item A and item B is perceivable by consumers.
