Scribd
Upload
51 views

Measuring Responses - Nominal Data: (Latin: Nomen Name) : The Items

The document describes different methods for measuring sensory responses, including classification, grading, ranking, and scaling. Classification sorts items into nominal groups. Grading relies on expert evaluation and commercial scales. Ranking arranges items in order of intensity. Scaling involves measuring attribute intensity using numbers or words, with category scaling providing ordinal data, line scaling providing interval data, and magnitude estimation sometimes providing ratio data.

Uploaded by

Francisco Hernandez Campos
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
51 views

Measuring Responses - Nominal Data: (Latin: Nomen Name) : The Items

The document describes different methods for measuring sensory responses, including classification, grading, ranking, and scaling. Classification sorts items into nominal groups. Grading relies on expert evaluation and commercial scales. Ranking arranges items in order of intensity. Scaling involves measuring attribute intensity using numbers or words, with category scaling providing ordinal data, line scaling providing interval data, and magnitude estimation sometimes providing ratio data.

Uploaded by

Francisco Hernandez Campos
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 14

SENSORY EVALUATION

MEASURING RESPONSES

• Nominal data: (Latin: nomen = name): the items


examined are placed in two or more groups which
differ in name but do not obey any particular order
or any quantitative relationship.

• Ordinal data: (Latin: ordinalis = order): the


panelist places the items examined into two or
more groups which belong to an ordered series.

• Interval data: (Latin: inter vallum = space


between ramparts): panelists place the items into
numbered groups separated by a constant interval.

• Ratio data: Panelists use numbers which


indicate how many times the stimulus in question is
stronger (or saltier, or more irritating) than a
reference stimulus presented earlier.
SENSORY EVALUATION

The most frequently used methods of measuring sensory response to a


sample are, in order of increasing complexity:

• Classification: The items evaluated are sorted into groups which differ in a
nominal manner; example: marbles sorted by color.

• Grading: Time-honored methods used in commerce which depend on


expert graders who learn their craft from other graders; example: “USDA
Choice” grade of meat.

• Ranking: The samples (usually three to seven) are arranged in order of


intensity or degree of some specified attribute; the scale used is ordinal.

• Scaling: The subjects judge the sample by reference to a scale of numbers


(often from 0 to 10) which they have been trained to use; category scaling
yields ordinal data or sometimes interval data, line scales usually yield interval
data, and magnitude estimation, although designed to yield ratio data, in
practice seems to produce mixed interval/ratio data.
SENSORY EVALUATION
CLASSIFICATION TEST

In classification tests, the subjects are asked to select an attribute or


attributes which describe the stimulus. In a beverage test, for example,
subjects place a mark next to the term(s) which best describe(s) the sample:

The selection of sensory attributes and the corresponding definition of these


attributes should be related closely to the real chemical and physical
properties of a product which can be perceived. Adherence to an
understanding of the actual rheology or chemistry of a product makes the
data easier to interpret and more useful for decision making.
SENSORY EVALUATION
GRADING TEST

Grading is a method of evaluation much used in commerce which depends


on expert “graders” who learn the scale used from other graders. Scales
usually have four or five steps such as “Choice,” “Extra,” “Regular,” and
“Reject.” Examples of items subjected to sensory grading are coffee, tea,
spices, butter, fish, and meat.

Grading systems can be quite elaborate and useful in commerce where


they protect the consumer against being offered low-quality products at a
high price, while permitting the producer to recover the extra costs
associated with provision of a high-quality product. However, grading
suffers from the considerable drawback that statistical correlation with
measurable physical or chemical properties is difficult or impossible.

Torry scale (Sanders and Smith, 1976)..


USDA scales (USDA, 1977).
SENSORY EVALUATION
RANKING TEST

In ranking, subjects receive three* or more samples which are to be


arranged in order of intensity or degree of some specified attribute. For
example, four samples of yogurt are to be ranked for degree of sensory
acidity, or five samples of breakfast cereal may be ranked for preference.

For each subject, the sample ranked first is accorded a “1,” that ranked
second a “2,” and so on. The rank numbers received by each sample are
summed, and the resulting rank sums indicate the overall rank order of the
samples. Rank orders cannot meaningfully be used as a measure of
intensity, but they are amenable to significance tests.

Chi cuadrada
Friedman
SENSORY EVALUATION
SCALING TEST
Scaling techniques involve the use of numbers or words to express the
intensity of a perceived attribute (sweetness, hardness, smoothness) or a
reaction to such attribute (e.g., too soft, just right, too hard). If words are
used, the analyst may assign numerical values to the words (e.g., like
extremely = 9, dislike extremely = 1) so that the data can be treated
statistically

The validity and reliability of a scaling technique are highly dependent upon:

• The selection of a scaling technique that is broad enough to encompass


the full range of parameter intensities and also has enough discrete points to
pick up all the small differences in intensity between samples;

• The degree to which the panel has or has not been taught to associate a
particular sensation (and none other) with the attribute being scaled; and

• The degree to which the panel has or has not been trained to use the scale
in the same way across all samples and across time
SENSORY EVALUATION
Although the properties of data obtained from any response scale may vary
with the circumstances of the test (e.g., experience of judges in the test,
familiarity of the attribute), it is typically assumed that:

• Category scaling (ISO term: rating) yields ordinal (or interval) data;

• Line scaling (ISO term: scoring) yields interval data;

• Magnitude estimation scaling (often called ratio scaling) sometimes, but not
always, yields ratio data.
SENSORY EVALUATION
CATEGORY SCALING
A category (or partition) scale is a method of measurement in which the
subject is asked to “rate” the intensity of a particular stimulus by assigning it a
value (category) on a limited, usually numerical, scale.

Category scale data are generally considered to be at least ordinal level data.

Generally do not provide values which measure the degree (how much) one
sample is more than another.
SENSORY EVALUATION
Generally, even word category scales are converted to numbers. The
numbers used in the above list are typical of the conversions which are
made.
SENSORY EVALUATION
LINE SCALING

With a linear or line scale, the panelist “rates” the intensity of a given stimulus
by making a mark on a horizontal line which corresponds to the amount of the
perceived stimulus.

The lengths most used are 15 cm and 6 in. with marks (“anchors”) either at the
ends or in 1.25 cm from the two ends.
SENSORY EVALUATION
MAGNITUDE ESTIMATION SCALING OR FREE NUMBER MATCHING
Is a scaling technique based on Stevens’ law. The first sample a panelist
receives is assigned a freely chosen number. (The number can be assigned
by the experimenter, in which case it is referred to as a modulus; or the
number can be chosen by the panelist.) Panelists are then asked to assign
all subsequent ratings of subsequent samples in proportion to the first
sample rating. If the second sample appears three times as strong as the
first, the assigned rating should be three times the rating assigned to the first
stimulus. Panelists are instructed to keep the number ratings in proportion to
the ratios between sensations. Examples:
SENSORY EVALUATION
SENSORY EVALUATION
SENSORY EVALUATION

You might also like