CONSUMER SURVEY DATA ENTRY AND CALCULATION

Various questions are suggested in a modadic or compared manner:

QUESTIONNAIRE AND DATA ENTRY

Questionnaire definition

Questionnaires are designed with the different types of questions. These questionnaires will be then displayed to the panelists very easily question after question.

Other parameters define product codes, experimental design, display screens, and so on, allowing a very user friendly and complete information during evaluation and consumer surveys.

Data Entry

Classical preparations or net preparations are suggested in order to enter automatically data a PC in an evaluation room or directly on net at home,

These preparations can also be used to create Word© paper forms collected directly by a scanner, or also, possibly, sent by mail and filled up and sent back for automatic data entry.

Other data entries can be suggested, for example by organizers, or also by Excel© imports,

or by direct data input on PC.

CALCULATION & ANALYSIS

For questions on preference scales, it is possible to edit many statistics: Means, standard-deviation, comparison tests: Student, Friedman, …, and also graphs, analysis of variance or multidimensional techniques as PCA, CFA... Automatic transformations from marks to ranks with ranking test, and distribution test with difference tests (Q of Cochran) are also available

For crossed tables, counting, percentages and graphs can be displayed according various analyses.

A text analysis module is also available to automatically count words, to take into account synonymous and word meanings in the sentence, and then, to issue various graphs and analysis.

More advanced techniques are also available to link Sensory Analysis data with the consumer preferences in order to focus on the product improvement studies (External Preference Mapping, Joint Principal Component Analysis).

Various calculation parameters can be defined, concerning the products selection, the variable selection, the test level, the panel selection, ...

General editions

Results can be previewed, saved in text file, printed directly on printer, or also built automatically in WORD© format.

Scale calculation: Graphs

Many graphs are then available, for example to describe results in histogram, in spider graph, in pair comparisons, by product, frequencies on multiple choices, on crossed table questions, on marks distribution, ...

All these graphs can be created again automatically with Excel©, using TASTEL macros dedicated to read tastel data and to display graphs in Excel©.

Various NONPARAMETRIC TESTS are also available:. On ranking, by transformation of intensity values in ranks, then with a Friedman test on it, . On normalisation validation, with Chi-square test and Kolmogorov-Smirnov tests, and, . On data distribution, with value groups, then Cochran Q test, ...

Various analysis can be conducted, Analysis of Variance, Principal Components Analysis,Discriminant Analysis, STATIS Analysis (closed to General Procrustean Analysis), Correspondence Factorial Analysis...

Analysis of Variance:

This analysis can be carried out with two, three, but also four factors, with the totality of linked interactions, analysis of normality are provided on the basis of the chi square test, then different mean comparison tests are available: Least Square Difference - Newman & Keuls test / Tukey test - Duncan test. Different reports can be issued to get summarised and detailed results, as well as roundups by segments of meaning.

Multidimensional Analysis

Principal Components Analysis:Three types of PCA can be conducted: PCA on Products (on product means), PCA on Assessors (on assessor means), or also PCA on Products X Assessors (on product and assessor values). Graphs in three dimensions can be displayed, as well as calculation values to help the analysis interpretation on three axis or on all axis.

This analysis will not be based on values, but will work on the value frequency according to the chosen scale. Graphs and figures associated with this technique, are similar to a PCA, but are working on frequency tables.

internal Preference Mapping :

This analysis details the consumer preferences in order to assess the behaviour preference diversity. Thus, this will allow representation of consumer segments based on 5 or more products. Graphs will display names, or panellist criteria, but also, it will be possible to preview the consumer densities, and so, the most interesting points to study.

Text analysis

Text analysis is dedicated to analyse free and suggested comments on products with possible several analysis steps as: Word filter, text descriptive analysis, synonym dictionary, and context definitions and results.

This analysis on the consumer comments allows a deeper analysis with a more qualitative product study.

Links between words can be defined to gather equivalent terms (synonymous) ; Meaning contexts can be also precised in order to distinguish positive or negative word expression, for example, but these differences can be based on other axis, for example between qualities and defects, or between olfactory aspects and gustatory aspects, and then counting results for each product and for each free question can be expressed.

After having analysed all the product of the study, it is possible to edit results in graphs (CFA or histograms) by selecting “CFA”, or “Frequency Graph”.

Preference Analysis - Probabilistic Analysis

This analysis is especially dedicated for protocols in Balance Incomplete Blocks, to find from personal ranking in block: "panellist", a collective order for the totality of the panel.

These models are developed for pairs and trios, extensions for quartets, ... are under way.

These calculations are based on frequencies to be first, second, or third.

In case of trios, two models are suggested:

Dependent model : Choice twice of the first product (frequency of the p product to be first to the square) by the choice once to be second (frequency of the p product to be second) - The whole, divided by the totality of terms, by forbidding the equal places.

Independent model : Choice of the first product among the three, then the second one among the two remainders.

Preference Analysis - Links with Sensory Analysis / Ideal product

Multidimensional techniques as “Joint PCA” can represent simultaneously consumer groups according to their appreciation on products, with expert evaluation on these same products. Indeed, an automatic link exists with our sensory analysis module. A preference optimum is calculated by the system, calculations are based on consumer answer densities. Then, with the help of this optimum, and with techniques of 'REVERSE INGENEERING', it is possible to find the sensory co-ordinates that a potential product would have had being placed in this place.

Checking the model is easily conducted by placing the sensory co-ordinate of this supplementary individual in the joint PCA: This position is the point 'PRODUIT IDEAL' (Ideal product).

These modelled co-ordinates are displayed in the last figure. These different statistical tools allow thus to bring rules of decision on sensory improvements to bring to products to get the best appreciation according to the concerned target.

Panel Management

In addition to calculations, panel management is available:

Panelist characteristics are followed up as references (age, occupation, ...), availability dates, or still consummation criteria or particular purchasing usages.

Performance and attendance can be issued by the System.

Segmentations and different targets can be taken into account for each panelist.

Data entry of this criteria can be done automatically by questionnaires, or also imported from Excel©.