Difference from Ideal

Difference from Ideal

Purpose

Analyses data from consumer descriptive profiling where there is a reference sample against which the other samples are evaluated.

Data Format

  1. Perfumes.xlsx
  2. Every second attribute column is the reference for the previous one. Therefore there must be an even number of attribute columns in the data set. For example, in the Perfumes.xlsx data set, Q1 – Intensity is referenced by Q2 – id Int.

Background

Analyses data from consumer descriptive profiling, where there is a reference sample against which the other samples are evaluated.
The reference profile may either be set to zero and the test samples profile given relative to these zero references, thus both plus and minus vales will be recorded.
Alternatively, the reference sample profile values are placed on the attribute scales and the test samples are evaluated relative to that reference profile.  In this event all ratings will be positive values.

Options

  1. Type: Type of means i.e. Adjusted or Arithmetic. Normally the adjusted means will be selected if there are missing data.
  2. Comparison:
    1. ‘Ideal-Intensity’ refers to subtracting the Intensity scores for each product from the Ideal Intensity. 
    2. ‘Intensity-Ideal’ refers to subtracting the Ideal scores from the Intensity of each product.
  3. Number of Decimals for Values: Required number of decimals for values given in the results.

Results and Interpretation

  1. The Difference tab shows the comparison values for each attribute and each product. These means values are calculated either as Ideal – Intensity or Intensity – Ideal depending on the option selected. The product means are calculated either as arithmetic means, or adjusted means according to the option selected. Adjusted means are calculated using the model attribute ~ product + assessor. In the Differences Plot products can be removed or added to the plot by selecting the product in the plot legend.
  2. The Names Check tab is used to check that the list of attributes and comparison attributes has been read in as expected. 

Technical Information

  1. The averagetable function from the SensoMineR package is used to calculate the adjusted means.
  2. Differences from Ideal are calculated manually.

References

Thierry Worch, Sébastien Lê, Pieter Punter, Jérôme Pagès.  Ideal Profile Method (IPM): The ins and outs. Food Quality and Preference. Volume 28, Issue 1, April 2013, Pages 45-59.