Choose the right test for your needs
In sensory analysis there are about 25 different test patterns. Like the tools in a mechanic’s toolbox, each test pattern is used to obtain a certain result. The tests patterns define the analytical procedures and statistical foundations for the achievement of the chosen objective.
Each tests differs from the other in complexity, number of samples to be analysed, number of samples served to each judge, skills and number of judges required for the test.
But how can we choose the right test for our needs? We can say that sensory tests can be grouped in categories or families. The tests in the same category answer to the same type of question:
- Discriminant tests: is there a difference between the samples?
- Directional tests: do the samples differ for a given characteristic (attribute)?
- Quantitative tests: how big is the difference between the samples?
- Descriptive tests: what attributes characterize the sample?
- Dynamic tests: how attributes evolve over time
- Affective tests: what is the level of satisfaction of the sample?
- Emotional tests: what emotions are evoked by the sample?
So as Panel Leaders, once we choose the right family, we can go on to select the specific test we think we should use. This decision is based on a series of factors such as the purpose of the analysis, the number and quantity of samples available, the sensory fatigue generated by the sample, the number of judges available and their level of expertise.
Discriminant tests will point out if there is a difference or similarity between samples. These tests are defined as non-specific, meaning that samples are compared and discriminated on the basis of their general characteristics and not on a specific attribute.
These tests are fairly simple to perform and do not require special skills by the panel. For this reason, in order to perform well, they need to involve a relatively high number of judges compared with other types of tests.
They are quick and easy to implement, making them an ideal choice for sample screening or routine analysis. Most of the tests in this category allow to compare only two samples at a time.
The better-known test in this category is the Triangular test; during a triangular test, three samples are served to each judge. Two of them are identical and one is different. The judges are asked to indicate the sample perceived as different. Generally, this test is conducted with the forced choice method, in the sense that the judge is asked to indicate a random sample if he does not perceive any difference between the samples. There must be an answer for each judge.
On the other hand, in directional tests comparison between samples focuses on a specific attribute (e.g. sweet, bitter, etc.). These tests are also quite simple and quick to perform and require limited skills from the judges. In addition to analytical attributes, hedonic attributes such as preference can be taken into consideration. Among the most used tests in this category we find the ranking test, during which the judges are asked to sort the samples according to the attribute studied (e.g. from the saltiest to the less salty).
In addition to detecting possible differences between the samples, the quantitative tests also allow to rate them on a scale. To use these tests the judges must be instructed beforehand. The tests belonging to this category allow you to compare multiple samples simultaneously. Among the most common we include the Difference vs. Reference test, which is a particularly useful during quality control. The judges are asked to rate on a scale the magnitude of the difference for each tested sample compared to a reference sample. While there are some exceptions, in general the extent of the difference is assessed in general terms. For this reason, this test falls in the “non-specific” type.
Descriptive tests are considered among the most important in the field of sensory sciences as they allow the identification of the distinctive characteristics (attributes) of a given product. The information that these tests provide is very “powerful” but, on the other hand, they are considered more complex than other types, both in terms of knowledge necessary for the panel leader to set up the test and for the time required to train the judges to recognise and rate the attributes in a similar way (panel calibration). For this reason, in recent years, research has devoted a lot of effort to developing “rapid” descriptive tests, which allow to determine the descriptive characteristics of a product without a trained panel. The CATA (Check All That Apply) test is one of the quick methods. It requires the judge to pick from a list of predefined attributes, the characteristics that best describe the product under evaluation.
Dynamic tests are more complex as they allow to determine the evolution of attributes during the evaluation of the product. These tests can only be conducted using an IT tool since the evolution of the attributes must be recorded as a function of the time spent during the evaluation of the sample. The best known of these tests is the Temporal Dominance Sensation (TDS); with this test the judges select the dominant attributes (e.g. sweet, bitter, salty) as they evolve during the product consumption (i.e. for food products, during mastication).
With hedonic tests we leave the field of analytical tests and enter that of affective tests. The tests in this category allow you to know the preference score of a product. They have a subjective character and for this reason they must be conducted on a large number of judges, generally identified among the consumers of the product under analysis. The most common of these tests is the acceptability test, in which the judges have to indicate the level of satisfaction related to the sample on a scale. In these tests, the judges can be either asked to express a generalised evaluation or to rate one or more particular attributes ( e.g. appearance, smell, taste etc … ) of the sample.
Emotional tests alsotake into consideration the subjective opinion of the consumer. These tests allow to understand the emotions that are elicited during the evaluation of the products. The collection of emotional responses is generally conducted through the compilation of a questionnaire. These tests do not require particular skills from the judges and are quite simple to perform. On the other hand, they require specific skills from the panel leader for the drafting of the questionnaires and the interpretation of the results. The EmoSemio test is among the latest tests developed for this category. It asks the judges to choose from a predefined list the (emotional) terms aroused by the product in question. It differs from other tests because it requires the use of a semiotic methodology for the choice of emotional terms and for drafting the questionnaire
The over 20 tests described in the scientific literature can fall into one or more of the categories illustrated above. We pictured these categories as a subway map. Each line will take you to a final destination, that is, the question you want to answer. The specific tests can be considered metro stations that provide access to one or more specific questions at the same time. Here is how the map should look.