Figure 4
The two horizontal bar graphs are arranged side-by-side. Each graph consists of two horizontal axes. The bottom horizontal axis is labeled “Importance”, and the top horizontal axis is labeled “Normalized Importance”. The top horizontal axis ranges from 0 percent to 100 percent in increments of 20 percent. The left bar graph is titled “Model 1”: The bottom horizontal axis ranges from 0,00 to 0,25 in increments of 0,05 units. The vertical axis lists the categories from top to bottom: “age”, “p 31”, “p 2”, “discount”, and “product”. There are 5 bars in the graph. The data is as follows: age: Normalized Importance: 104.2 percent. Importance: 0,263. p 31: Normalized Importance: 90.6 percent. Importance: 0,238. p 2: Normalized Importance: 67.5 percent. Importance: 0,178. discount: Normalized Importance: 65.3 percent. Importance: 0,172. product: Normalized Importance: 60 percent. Importance: 0,151. The right bar graph is titled “Model 2”: The bottom horizontal axis ranges from 0,0 to 0,3 in increments of 0,1 units. The vertical axis lists the categories from top to bottom: “age”, “p 2”, “p 31”, “product”, and “discount”. There are 5 bars in the graph. The data is as follows: age: Normalized Importance: 96.7 percent. Importance: 0,290. p 2: Normalized Importance: 69.7 percent. Importance: 0,209. p 31: Normalized Importance: 63.7 percent. Importance: 0,191. product: Normalized Importance: 60.7 percent. Importance: 0,182. discount: Normalized Importance: 42.7 percent. Importance: 0,128. Note: All numerical values are approximated.Independent variables normalized importance. P31 = Internet usage; p2 = purchase frequency. Source: Own elaboration. SPSS multilayer perceptron