With SPSS Categories, you get the Leiden University family of scaling procedures, including correspondence analysis with a graphical user interface.
Whatever types of categories you study -- market segments, political parties or biological species -- SPSS Categories' optimal scaling procedures free you from the restrictions of two-way tables, placing the relationships among your variables in a larger frame of reference. See a map of your data -- not just a statistical report.
SPSS Categories: What It Is
- With quantitative data, researchers typically employ scatterplots to understand their data. Using SPSS Categories, researchers can produce plots for their categorical data.
- SPSS's first upgrade of the Categories option to SPSS Base.
- Features two new procedures -- A greatly enhanced correspondence analysis procedure (the best there is) and a procedure for "categorical regression analysis," -- along with three other procedures.
- The two new procedures have output that displays as tables and charts in the SPSS Viewer.
- Altogether, provides a framework for multivariate analysis of data of mixed measurement levels.
Who Uses It
- Market research
- Survey research
- Social and behavioral science
- Educational research
- Medical and health science research
- Seriation or ordination of archaeological or historical data
- Anyone with two-way or multi-way tables, especially large, unwieldy ones
- Anyone with ordinal data. Examples:
- Customer satisfaction studies - How satisfied are you? Very satisfied, somewhat satisfied, etc.
- Medical - Patient recovery: Well, mildly impaired, moderately impaired, severely impaired.
- Behavioral - How effective was therapy: Poor, fair, good, very good, excellent.
- Anyone with aggregate data collected at multiple time points.
What It Helps You Do
- Understand the information in large two-way and multi-way tables.
- Work with and understand ordinal and nominal data in procedures analogous to conventional regression, principal components, and canonical correlation.
- Visualize and explore multivariate categorical data.
How It Works
- CORRESPONDENCE performs a singular value decomposition of a transformed two-way array of numbers. This enables the user to represent the information in the array in a low-dimensional approximation. The output includes information that can be rendered in a biplot, which produces revealing insights into the data.
- HOMALS performs multiple correspondence analysis of two or more categorical variables treated as if the categories are unordered. This can be viewed as a type of principal components analysis of categorical variables. The output includes a plot of the category quantifications, which gives insight into the relationships between the variables in the analysis. The category quantifications are centroids of the objects that share that category. When viewed in the same space, the category quantifications and the object scores comprise one type of biplot.
- PRINCALS, OVERALS, and CATREG are analogous to principal components, canonical correlation, and regression. All three procedures allow the user to specify a transformation type of nominal, ordinal, or numeric, on a variable-by-variable basis. All three procedures use an alternating least squares algorithm.