In sensory and consumer studies the aim is usually to compare products for each attribute tested. This module reports summary statistics for each pair of products in the data and is a useful starting point for understanding how the assessments of products differ.
For each pair of products, eg A and B, the number of assessments where A scores more than B, A equals B, and where A scores less than B are reported alongside the average difference in scores (measure of the ‘signal’) and the standard deviation of the difference (measure of the ‘noise’). Therefore the comparison is made of the two products and then summarised rather than the usual approach of summarising and then comparing the averages.
For EyeOpenR to read your data, the first five columns of the ‘Data’ sheet must be in the following order: assessor (consumer), product, session, replicate and order (sequence). For sensory analysis the data for attributes should be in the sixth column (column F) onwards. There should be one column for each attribute. The attributes data should be numeric. For consumer analysis the data for consumer liking and other consumer assessments (ratings) should be in the sixth column (column F) onwards. There should be one column for each rating. These are described as attributes in the options.
If there is no session, replicate or order information then these columns should contain the value ‘1’ in each cell.
Additional information about the data in the
‘Data’ sheet can be included in additional sheets. The ‘Attributes’ sheet can
be used to specify the names of the attributes, data types and minimum and
maximum values that are used to check data quality. The ‘Assessors’ sheet can
be used to specify assessor names if codes are used in the ‘Data’ sheet.
Similarly, the ‘Products’ sheet can be used to specify product names if codes
are used in the ‘Data’ sheet. See the example spreadsheet for an illustration
of the data format.
This module does not test whether differences between products are statistically significant, but it does report the standard deviation of the mean differences to help contextualise the size of the difference relative to the variation.
Comparison: There is one table for each pair of products in the data, these are selected by clicking on the box that describes the pair, eg Product 1 and Product 2. This shows a table with the attributes in the rows and five columns that describe the difference assessments of the pair of products for the attribute.
The first column reports the number of assessments (assessor by session by replicate) where the first product was given a lower value for the attribute than the second product. The second column reports the number of times they received the same assessment. The third column reports the number of times the first product was given a higher value for the attribute.
The fourth column shows the average difference between the assessments. This can give additional information when the difference between the products is consistently in one direction. For example, if there were 10 assessments and in all 10 cases the first product had a higher rating then you would know that this assessment was consistent. The means would then describe whether the difference was consistent but small or consistent and larger.
The fifth column shows the standard deviation of the difference between the assessments. This helps to put the mean difference into context. If the mean difference is considered the signal of a difference then this is the noise and the size of the difference relative to the standard deviation is an interesting measure of how meaningful the size of the mean is relative to underlying variability in assessments.
Note that the results for Product 2 and Product 1 would be the same as those for Product 1 and Product 2 and are therefore not shown to make the output manageable. However, the mean value would have a different sign, so if it is positive for Product 2 and Product 1 it is negative for Product 1 and Product 2. Having the first three columns helps to immediately see which product is attracting the higher ratings.
R packages used: