Descriptive Analysis Workbook Scale Mean Usage
What is it?
Scale Mean Usage is one of the sheets in the Descriptive Analysis Workbook . A table
with the mean usage scores based on a logarithmic scale for each panelist. The calculation
takes place across all samples, sessions, and attributes.
Why would I use it?
Find out which panelists are tending to scale attributes lower or higher than the rest of the panel on the scale.
Setup options
To generate the Scale Mean Usage sheet, under 2. Select options , select Panelist performance summary . Update the thresholds as required. |
Analysis Overview
Using
the log allows us to compare the relative difference since the means will vary
from attribute to attribute so it doesn't make sense to use a linear scale.
- Panel Mean Usage . The log of the ratio of the panel mean
for a specific attribute to that of all attributes.
A positive value occurs when the
grand attribute mean is greater than the grand mean of all attributes.
-
Attribute has higher intensities compared to the mean of all attributes.
A negative value occurs when the
grand attribute mean is less than the grand mean of all attributes.
- Attribute
has lower intensities compared to the mean of all attributes.
The larger the value, the bigger
the difference between the two means.
- Panelist Mean Usage . The
log of the ratio of the panelist mean for a specific attribute to that of panel mean for that attribute.
A
positive value occurs when the attribute panelist mean is greater than the
attribute grand mean.
- Panelist is scaling attribute higher than panel mean
for the attribute.
A
negative value
occurs when the attribute panelist mean is less than the attribute grand mean.
- Panelist is scaling attribute lower than panel mean for the attribute.
The
larger the value, the bigger the difference between the two means.
- Distance of the Panelist Mean Usage (number inside bracket ) . Distance of the panelist scale mean usage
from the average of all panelist scale mean usages by attribute.
A
positive value occurs when the panelist mean usage is greater than the average
of all panelist scale mean usages.
- Panelist is scaling attribute higher than panel
mean for that attribute.
A
negative value
occurs when the panelist mean usage is less than the
average of all panelist scale mean usages.
-
Panelist is scaling attribute lower than panel mean for that attribute.
The
larger the value, the bigger the difference between the scale mean usages
- Highlighted cells . Distance of the panelist scale mean usage from the panel scale mean usage by
attribute is greater than the preset threshold.
A
difference of 10, 20 and 30 represents 2x, 4x and 8x the panel mean.
The
yellow highlighting quickly identifies panelists using 2x outside of the panel
mean for all attributes. The number tells you in which direction and how large
the difference is. The number in brackets gives you an idea of the scale usage
difference within the specific attribute.
- Percentage . The number of attributes below the preset threshold/ the total number of attributes *100.
Scale Mean Usage Definitions
=log(Attribute Grand
Mean/Grand Mean of all Attributes,2)*10
A NegInf value occurs if
there is an attribute scored with the value of 0 for all samples across all
sessions, which provides an infinitely negative response. Due to this when
the Average of Panelist Mean Score for all panelists for the
attribute is calculated to determine the per panelist value in
brackets, the calculation results in a PosInf value for all panelists for this
attribute.
Example Conclusion
Noel Crabtree is outside the
threshold meaning he is 2x
outside the panel mean for
Fresh Apple Flavour
.
Panelist mean usage is negative so
we know he is scoring lower.
From
the panel mean usage we know that Fresh Apple Flavour has a higher intensity
compared to the mean of all attributes.
We
can see that he is using the scale differently for the same attributes that
were picked up in the crossover but these attributes fall within the mean.
Brown Colour and viscosity are highlighted for all panelists. This suggests that
these attributes are scored so low they likely do not play a roll in the
differentiation. As we can imagine, viscosity would not vary too much as they
are all juices being evaluated.
Related Articles
Descriptive Analysis Workbook Scale Range Usage
What is it? Scale Range Usage is one of the sheets in the Descriptive Analysis Workbook . A table with the mean range scores based on a logarithmic scale for each panelist. The calculation takes place across all samples, sessions, and attributes. ...Descriptive Analysis Workbook Panelist Mean Scores
What is it? Panelist Mean Scores is one of the sheets in the Descriptive Analysis Workbook . One table per attribute with different indicators of discrimination among panel and panelists. Why would I use it? Further investigate discrimination ability ...Descriptive Analysis Workbook Review
Overview The Descriptive Analysis Workbook provides detailed analyses on product and panelist performance. The report is compatible with sample related Line Scale , Category , LMS , Numeric , and ' Choose Only 1 ' questions when 2 or more samples ...Descriptive Analysis Workbook Panelist Mean Graphs
What is it? Panelist Mean Graphs is one of the sheets in the Descriptive Analysis Workbook . It is a visual overview of the product intensities for each attribute, showing the panel and panelist means. Why would I use it? It can be helpful for ...Descriptive Analysis Workbook Product Mean Graphs
What is it? Product Mean Graphs is one of the sheets in the Descriptive Analysis Workbook . It is a visual representation of panel and panelist product intensities for each attribute. Why would I use it? It can be helpful for visually comparing ...