User manual SPSS CATEGORIES 11.0

[. . . ] SPSS Categories 11. 0 ® Jacqueline J. Heiser SPSS Inc. For more information about SPSS® software products, please visit our Web site at http://www. spss. com or contact SPSS Inc. 233 South Wacker Drive, 11th Floor Chicago, IL 60606-6412 Tel: (312) 651-3000 Fax: (312) 651-3668 SPSS is a registered trademark and the other product names are the trademarks of SPSS Inc. No material describing such software may be produced or distributed without the written permission of the owners of the trademark and license rights in the software and the copyrights in the published materials. [. . . ] This data set can be found in coffee. sav. For each of twenty-three iced coffee image attributes, people selected all brands that were described by the attribute. Table 11. 1 contains the attributes and their corresponding labels. The six brands are denoted AA, BB, CC, DD, EE, and FF to preserve confidentiality. Table 11. 1 Iced coffee attributes Image Attribute Label Image Attribute Label good hangover cure low fat/calorie brand brand for children working class brand rich/sweet brand unpopular brand brand for fat/ugly people very fresh brand for yuppies nutritious brand brand for women minor brand cure low fat children working sweet unpopular ugly fresh yuppies nutritious women minor fattening brand appeals to men South Australian brand traditional/old fashioned brand premium quality brand healthy brand high caffeine brand new brand brand for attractive people tough brand popular brand fattening men South Australian traditional premium healthy caffeine new attractive tough popular The setup of the data requires that the cases be weighted by the variable freq. To do this, from the menus choose: Data Weight Cases. . . Weight cases by Frequency variable: freq 172 Chapter 11 Principal Normalization Initially, you will focus on how the attributes are related to each other and how the brands are related to each other. Using principal normalization spreads the total inertia once over the rows and once over the columns. Although this prevents biplot interpretation, the distances between the categories for each variable can be examined. Dimensionality In order to decide how many dimensions to use, you can find an initial solution in five dimensions and choose a number of dimensions that accounts for the bulk of the inertia. To obtain an initial solution in five dimensions with principal normalization, from the menus choose: Analyze Data Reduction Correspondence Analysis. . . Row: image Column: brand Select image. Category range for row variable: image Minimum value: 1 Maximum value: 23 Click Update. Category range for row variable: image Minimum value: 1 Maximum value: 6 Click Update. Dimensions in solution: 5 Normalization Method Principal Figure 11. 21 shows the decomposition of the total inertia along each dimension. Two dimensions account for 83% of the total inertia. Adding a third dimension adds only 8. 6% to the accounted for inertia. Thus, you elect to use a two-dimensional representation. Correspondence Analysis Examples 173 Figure 11. 21 Inertia per dimension Proportion of Inertia Inertia . 506 . 159 . 069 . 055 . 015 . 804 3746. 968 Chi Square Accounted for Cumulative . 629 . 198 . 086 . 068 . 018 1. 000 . 629 . 827 . 913 . 982 1. 000 1. 000 Dimension 1 2 3 4 5 Total Singular Value . 711 . 399 . 263 . 234 . 121 To compute a two-dimensional solution, recall the Correspondence Analysis dialog box: Model. . . Scatterplots Row points Column points Contributions Figure 11. 22 shows the contributions of the row points to the inertia of the dimensions and the contributions of the dimensions to the inertia of the row points. If all points contributed equally to the inertia, the contributions would be 0. 043. Healthy and low fat both contribute a substantial portion to the inertia of the first dimension. Men and tough contribute the largest amounts to the inertia of the second dimension. Both ugly and fresh contribute very little to either dimension. 174 Chapter 11 Figure 11. 22 Attribute contributions Contribution Of Point to Inertia of Dimension Of Dimension to Inertia of Point 1 . 652 . 512 . 114 . 454 . 296 . 953 . 702 . 893 . 911 . 404 . 771 . 446 . 941 . 179 . 693 . 212 . 585 . 000 . 196 . 392 . 946 . 965 . 593 2 . 173 . 480 . 152 . 260 . 509 . 020 . 053 . 021 . 007 . 560 . 003 . 209 . 021 . 380 . 255 . 368 . 085 . 131 . 214 . 246 . 006 . 007 . 007 Total . 825 . 992 . 266 . 715 . 805 . 973 . 755 . 914 . 918 . 964 . 774 . 655 . 962 . 559 . 948 . 580 . 670 . 131 . 410 . 637 . 951 . 972 . 600 IMAGE fattening men South Australian traditional premium healthy caffeine new attractive tough popular cure low fat children working sweet unpopular ugly fresh yuppies nutritious women minor Active Total 1 . 042 . 073 . 010 . 039 . 016 . 152 . 019 . 086 . 035 . 056 . 058 . 008 . 175 . 006 . 055 . 020 . 011 . 000 . 001 . 010 . 041 . 062 . 027 1. 000 2 . 035 . 219 . 044 . 071 . 090 . 010 . 005 . 006 . 001 . 246 . 001 . 011 . 013 . 041 . 064 . 112 . 005 . 002 . 002 . 019 . 001 . 001 . 001 1. 000 Two dimensions contribute a large amount to the inertia for most row points. The large contributions of the first dimension to healthy, new, attractive, low fat, nutritious, and women indicate that these points are very well represented in one dimension. [. . . ] Intra-individual and inter-individual multidimensionality. In: Psychological Scaling: Theory & Applications, H. Nonlinear canonical correlation and some related techniques. 316 Bibliography Leiden: DSWO Press. British Journal of Mathematical and Statistical Psychology, 36: 54­80. [. . . ]