Individual Spaces

An individual space is computed for each source. The dimension weights show how the individual spaces load on the dimensions of the common space. A larger weight indicates a larger distance in the individual space and thus greater discrimination between the objects on that dimension for that individual space.

Specificity is a measure of how different an individual space is from the common space. An individual space that was identical to the common space would have identical dimension weights and a specificity of 0, while an individual space that was specific to a particular dimension would have a single large dimension weight and a specificity of 1. In this case, the most divergent sources are Breakfast, with juice, bacon and eggs, and beverage, and Snack, with beverage only.
Importance is a measure of the relative contribution of each dimension to the solution. In this case, the dimensions are equally important.

The dimension weights chart provides a visualization of the weights table. Breakfast, with juice, bacon and eggs, and beverage and Snack, with beverage only are the nearest to the dimension axes, but neither are strongly specific to a particular dimension.

Figure 3. Joint plot of individual space "Breakfast, with juice, bacon and eggs, and beverage"

The joint plot of the individual space Breakfast, with juice, bacon and eggs, and beverage shows the effect of this scenario on the preferences. This source loads more heavily on the first dimension, so the differentiation between items is mostly due to the first dimension.

Figure 4. Joint plot of individual space "Snack, with beverage only"

The joint plot of the individual space Snack, with beverage only shows the effect of this scenario on the preferences. This source loads more heavily on the second dimension, so the differentiation between items is mostly due to the second dimension. However, there is still quite a bit of differentiation along the first dimension because of the fairly low specificity of this source.