7.4 Conclusion

Rows Factor Scores:
○ Component 1: Viena and Pavo
○ Component 2: Salchicha pavo Chero VS Salchicha Viena Nutrideli

Cols Factor Scores:
○ Component 1: Martha (Floury & Spicy), Diana (Acidic)
○ Component 2: Lalo (Rubbery), Dulce (White Meat)

General Insights:
○ Sausage Type (Session) is the best grouping strategy for Rows.
○ Sausage Type provide best narrative and a common theme (vs PCA, PLSC).
○ Negligible differences between Panelists.

At this point, we can say that MFA provides the most comprehensive picture about our sausage dataset compared to PCA, BADA, DicA, MCA, PLSC. MFA can present all of the insights found by each of the methods presented throughout this book while saving analyst from the headache of centering/scaling because MFA can control for the bias among panelists. From MFA, we have re-discovered and confirmed the relationships among observations and the differentiation between Pavo vs Viena; only this time, we can also assess the effect of each Panelist. We can see that there are negligible differences between Panelists. The Panelist that constantly disagree with others is Diana in which she is more sensitive to Acidic and Bitter. We also noted that sausages of Pavo variants tend to have larger differences between Panelist. For vaiables, we can see the presence of the same set of notable variables that we have seen in previous methods including White Meat, Dark Meat, Floury, Rubbery, Bitter, Acidic, Spicy, Umami & Co. Here we suggest that we can interpret the realtionships between variables according to 3 main themes: Textures (Rubbery, Floury) VS Meat Types (Dark Meat, White Meat), Common sausage tastes (Umami, Smokey, Acidic) VS Pain (Spicy), and from PLSC - Most attributes can be related to the 5 Basic Tastes.

7.4.1 Save to PPTX:

# Here we can save all figures to a PowerPoint
savedList <- saveGraph2pptx(file2Save.pptx = 'AllFigures_MFA 12', 
                            title = 'All Figures for MFA', 
                            addGraphNames = TRUE)