IBM SPSS Statistics 31

Proximity mapping is a visualization technique that is used to reduce the dimensionality of multivariate data and to display relationships among objects (cases, items, or other entities) in a spatial configuration. It supports multiple proximity sources, incorporates additional variables (attributes and properties), and offers a wide range of transformation and restriction options. These features make PROXMAP a powerful tool for exploring multivariate structure and integrating diverse types of data into a unified spatial representation.

Distance Correlation is a versatile metric that detects any form of statistical dependence between variables (linear or nonlinear). It addresses the limitations of analyzing real-world data that often exhibits nonlinear and complex dependencies.

Time series filtering is a vital technique in econometrics and time series analysis, often employed to decompose a time series into trend and cyclical components. This is especially important in macroeconomics and finance, where identifying business-cycle fluctuations or long-term economic trends is essential. Time Series Filters (TSF) is added as an extension procedure with Python plug-in. Search and install the extension Time Series Filters from extension hub (Extensions > Extension hub).

The Conditional Inference Tree procedure estimates classification and regression trees, by using a unified framework for conditional inference or permutation tests. It provides more stable trees with reduced chances of overfitting. The two types of trees that are provided by this procedure are conditional inference trees and model-based trees. Conditional Inference Tree is added as an extension procedure with R plug-in. Search and install the extension Conditional Inference Tree from extension hub (Extensions > Extension hub).

STATS EARTH uses the MARS algorithm to perform regression analysis to detect nonlinearities and interactions. This is added as an extension procedure with R plug-in. Search and install the extension Multiple Adaptive Regression Splines from extension hub (Extensions > Extension hub).

Curated Help is a feature that analyzes the output of a procedure and provides a summary of key findings. For IBM SPSS Statistics 31, Curated Help is available for the following procedures. Bivariate Correlations (Analyze > Correlate > Bivariate), Partial Correlation (Analyze > Correlate > Partial), Distances (Analyze > Correlate > Distances), Canonical Correlation (Analyze > Correlate > Canonical Correlation), Correlation in Linear Regression (Analyze > Regression > Linear)

The Analyze menu now includes a direct option for Chi Square test of independence, and Independent-Samples T-Test features an optional control for homogeneity of variance testing. Additionally, coefficient of variation calculation has been implemented across multiple procedures, enabling more detailed statistical analysis.

Dark mode support has been extended to table headers, toolbar icons, charts, and the Output Viewer, providing a more comfortable working environment. Excel import now offers the option to read variable names from the first row, and the new output theme creation feature allows users to save and reuse preferred styles for pivot tables and charts. The ability to set background images in charts has also been added.

Users can now directly search for Design of Experiments procedures from the application search bar, and unlicensed features are displayed in the menu with upgrade guidance. Daily operations have been streamlined with features such as default column width settings in Variable View and the display of selected case counts when filters are applied.

Python has been upgraded to version 3.13.1, while Java Runtime Environment (JRE) and JDK have been updated to 17.0.13.0. Support for the new Chinese character standard (GB18030-2022) has been added. Furthermore, data import from Cognos Analytics to IBM SPSS Statistics is now possible, creating a more flexible data analysis environment.

Boruta Feature Selection compares the importance of original variables with shadow attributes that are created by shuffling case values of original variables. Variables less important are removed, more important ones are confirmed, and the rest are marked tentative until resolved. This is added as an extension procedure with R plug-in. Search and install the extension STATS_BORUTAFEATURES from extension hub (Extensions > Extension hub).

Curated Help is available for T-Test procedures and provides dynamic interpretations of results. Curated Help is available for the following T-Test procedures: One Sample T-Test, Independent Samples T-Test, and Paired Samples T-Test.

With the AI Output Assistant, you can select output and ask questions to Large Language models (LLMs) and get explanations related to the output. The feature is powered by IBM watsonx.ai. You should have a trial or paid account in IBM watsonx to use the feature. Go to the menu bar in output viewer and click Analyze output with AI tools to open the AI Output Assistant.

The following new shortcut keys are added in Syntax Editor: Go to next error (Ctrl+E / Command+E) and Go to previous error (Ctrl+Shift+E / Command+Shift+E). These shortcuts make it more efficient to check and fix syntax errors.

JRE and JDK are upgraded to version 17.0.16. This improves the stability and performance of Java-based features and enables the use of the latest Java security features.