JMP platforms and statistical methods

Graph Builder:
Line graph, dot plots, scatter plots, linear regression

Distribution:
Histograms, summary statistics

Tabulate:
Creating new data tables of summary statistics

Multivariate:
Correlations

Fit Y by X:
One-factor analysis of variance, multiple comparisons

Fit Model:
Two-factor analysis of variance

Clustering:
Hierarchical clustering

Objective

Use ANOVA, regression, and clustering to describe and compare consumers’ preference for perfumes based upon their aroma characteristics, along with the age and social activity of the consumer.

Problem statement

A cosmetics manufacturer has conducted a sensory study of its perfume line with the intent of understanding the preferences that different types of consumers have for its products and its aroma characteristics. It currently produces nine different perfumes but may use the results of this study to consider other perfumes to create.

For the first part of the study, the nine perfumes were evaluated by six sensory experts and scored across 11 aroma attributes. The blindfolded experts evaluated each of the perfumes twice, with the 22 sessions performed in a random order. The purpose of this part of the study was to produce a quantitative profile for each of the perfumes.

For the second part of the study, 477 consumers who use perfume on a regular basis were recruited. Each of these consumers brought home the nine different perfumes. They were asked to use each perfume on two separate occasions and to do so in a randomly assigned order. They ranked each perfume on a scale of 1-10 on how likely they would be willing to buy the perfume in the future. The consumers were classified into one of 24 different segments based upon their age and the amount of social activity they would engage in when typically using perfume.