SPSS Statistics 17.0 New capabilities What s New in SPSS Statistics 17.0 Recognizing the increasingly critical role of analytics in helping organizations reach their goals, SPSS Inc. has made significant improvements to the suite of powerful but easy-to-use statistical software products that had served the needs of tens of thousands of people in business, government, academia, and research since 1968. SPSS Statistics 17.0 provides new and enhanced capabilities to support you throughout the analytical cycle and ensure that your organization can more easily make decisions based on reliable data analyses. In this release, you ll find: n Improved tools for research and reporting n Easier ways for beginners and generalists to perform analysis, along with new functionality for advanced users and statistical programmers n Greater support for enterprise integration and the deployment and management of analytical assets All new features are available on the Microsoft Windows, Apple Mac OS, and Linux platforms.* And, in this release, those using SPSS Statistics on the Mac OS and Linux platforms can connect to SPSS Statistics Server, with appropriate licensing. This improves analytical performance, facilitates the management and deployment of analytic results, and provides IT with the necessary tools for integration, configuration, and maintenance of a client/ server deployment of SPSS Statistics. The SPSS Statistics Adapter is now also available to users of SPSS Statistics 17.0 on Mac OS and Linux, allowing organizations with mixed desktop environments to give all SPSS Statistics users access to the features and services available in SPSS Predictive Enterprise Services. (For more information about this advanced platform for managing, automating, and deploying analytical assets, go to www.spss.com/predictive_enterprise_services.) Improved tools for research and reporting Researchers benefit directly from the new research and reporting features available in SPSS Statistics 17.0. Multiple imputation of missing values One challenge survey researchers and others often encounter is missing data. Now, using SPSS Missing Values 17.0**, you can impute missing values in either categorical or continuous variables by multiple imputation. You ll obtain more accurate standard errors and significance levels than if data were imputed using a single imputation method, and gain a clearer picture than if you had ignored missing data in your analysis. Multiple imputation in SPSS Missing Values conducts imputation using the monotone or fully conditional specification methods. You can also have the most appropriate method selected automatically. Codebook creation In SPSS Statistics Base, survey researchers will be able to automatically create a codebook describing their dataset, eliminating the time-consuming task of creating one manually. The codebook contains dictionary information such as variable names, variable labels, value labels, missing values, and frequencies, allowing the user to easily communicate the results of a survey or questionnaire to anyone. * Amos 17.0, SPSS Viz Designer, and SPSS Exact Tests 17.0 are only available on the Microsoft Windows platform, and the exchange of data with SPSS Inc. s Dimensions family of survey research products is supported only on the version of SPSS Statsitics 17.0 that operates on Windows. **This module was formerly known as SPSS Missing Value Analysis.
More control when exporting to Microsoft Office New to SPSS Statistics 17.0 is more control at the time of export over wrapping and shrinking of tables for placement in Microsoft Office products; you can also export to existing Microsoft Excel spreadsheets. Export options can be controlled via the export interface in SPSS Statistics or via the syntax command language, allowing users to create automated export formatting for production jobs. Because you won t need to format SPSS Statistics tables manually in Office, you can create presentation-ready reports far more quickly. Nearest neighbor analysis This procedure enables analysts to use SPSS Statistics Base to quickly link cases to nearest neighbors to help group valued customers and identify competitors. It can be used in marketing, in clinical studies to select control cases that are similar to clinical cases, and in many other applications. Model Viewer integration provides users with greater insight into the analysis. Other enhancements to statistical procedures in this release include the ability to choose aggressive or conservative rounding in the COMPUTE procedure, which enables researchers to reproduce results from prior versions of SPSS Inc. software. Plus, refinements to a number of procedures in SPSS Statistics Base and in SPSS Advanced Statistics give users more flexibility in analyzing data. For details, see the comparison chart on pages 5-8. Enhancements to other SPSS Statistics modules Three new regularization methods have been added to SPSS Categories : Ridge Regression, the Lasso, and the Elastic Net. Regularization aims at improving predictive accuracy by stabilizing the parameter estimates and facilitates the analysis of high-volume data. SPSS Categories also offers three new predictive accuracy assessment methods: systematic multiple starts,.632(+) bootstrap, and cross validation (CV). The first method enables you to find a model that is optimal for describing the relations between variables; the other methods enable you to find those that are optimal for prediction. Together, these methods make it possible to create reliable predictive models from shallower-wider datasets. Improved research and reporting can also be achieved, using Amos 17.0, a product for structural equation modeling (SEM) that integrates with SPSS Statistics 17.0. Enhancements to Amos 17.0 include the ability to write a Visual Basic program based on a model specified using Amos Graphics; to copy and paste part of one path diagram to another; and to automatically set up parameter constraints when creating a latent growth curve model. More details on Amos 17.0 can be found at www.spss. com/amos. New functionality for generalists and for specialists In this release, there are features that make SPSS Statistics more accessible to beginners and generalists, and other capabilities that meet the needs of expert analysts and statistical programmers. RFM analysis With this release, SPSS Inc. introduces the SPSS EZ RFM module, which enables marketers to target campaigns more effectively by making the complexities of RFM analysis simple and straightforward. A custom interface guides marketers step by step, as they review the recency, frequency, and monetary value of customer transactions. Now even those without a strong statistical background can quickly gain clearer insight that will lead to more targeted, cost-effective and successful campaigns. And since marketers can perform this RFM analysis on their own, they can do so more frequently and consistently. (For more details on this new module, see www.spss.com/statistics/ez_rfm.) SPSS Advanced Statistics was formerly called SPSS Advanced Models. 2
Additional visualization options If you need to create customized graphs of your results, or want to save special graph types as templates for others to use, the Graphboard feature of SPSS Statistics 17.0 offers integration with a new product, SPSS Viz Designer. With this new functionality a user can apply chart templates created in SPSS Viz Designer to analyses run in SPSS Statistics. (For additional information, please go to www.spss.com/viz_designer.) Statistical programmers will find additional graphing options available in SPSS Statistics 17.0. This release extends support of the SPSS Statistics plug-in for R to include graphic packages written in that programming language. Since these can be saved as templates, analysts will be able to choose from an almost unlimited array of potential graph types when communicating their findings. Enhanced programmability SPSS has updated the plug-ins for Python, the.net version of Visual Basic, and R for compatibility with SPSS Statistics 17.0. The R plug-in for SPSS Statistics 17.0 also includes the ability to drive chart creation from R and have the results included in the SPSS Statistics Output Viewer for report creation. More information about the SPSS Statistics Programmability Extension, free plug-ins created by SPSS Inc., custom procedures created by other users, and the SDK can be accessed at www.spss.com/devcentral. More support at the enterprise level Because organizations are increasingly relying on predictive analytics to drive business decision making, SPSS Inc. has enhanced several key features in SPSS Statistics 17.0 that provide easier enterprise integration, as well as the deployment and management of analytical assets. Improved Syntax Editor Expert users of the SPSS Statistics suite will welcome new features in the SPSS Statistics Syntax Editor such as autocompletion, color coding of syntax, a gutter to display line numbers and break point, and more. These help you create SPSS Statistics syntax more quickly and easily and check for errors before the task is executed. By catching errors early, these enhancements will make both analysis and report delivery more efficient. Deeper integration with SPSS Predictive Enterprise Services SPSS Predictive Enterprise Services is a platform that supports organizations in the management of analytical assets, automation of analytical processes, and efficient delivery of results as they move toward becoming Predictive Enterprises. The deeper integration available in SPSS Statistics 17.0 streamlines the use of SPSS technologies throughout the enterprise. Custom Dialog Builder SPSS Statistics 17.0 introduces a Custom Dialog Builder so that more experienced users can make existing dialogs easier for business users, and create dialogs for custom features built through programmability. The Custom Dialog Builder enables your organization s less experienced users to quickly learn how to perform routine operations efficiently, and gives programmers a way to deploy their work efficiently. For example, if your organization uses products from more than one SPSS product family SPSS Statistics and Clementine, for example you can define a common data access interface once, and then use it in both analytical tools to ensure consistent results. The SmartViewer product is replaced in this release by the new SPSS SmartReader user interface. This provides much the same functionality. For instance, report consumers will be able to view and pivot SPSS Statistics output and retrieve files from SPSS Predictive Enterprise Services. 3
Enhanced administrative tools New administrative tools available in SPSS Statistics Server provide IT staff with greater control in configuring, monitoring, prioritizing, troubleshooting, and optimizing their SPSS Statistics client/server environment. Information now available includes activity by SPSS Statistics users, and resources used in the execution of SPSS Statistics analysis. Also, someone opening a data file that is already in use would be automatically alerted to this, enabling teams to work together more efficiently and effectively by knowing who is accessing data at any given time. SPSS Statistics Server is officially supported for deployment in virtualization environments, giving organizations greater control over the resources it has access to, and greater flexibility in their chosen client/server configurations. For more information on SPSS Statistics Server, please visit www.spss.com/ statistics/server. System requirements SPSS Statistics Base 17.0 for Windows n Operating System: Microsoft Windows P (32-bit versions) or Vista (32-bit or 64-bit versions) n Hardware: Intel or AMD x86 processor running at 1GHz or higher Memory: 512MB RAM or more; 1GB recommended Minimum free drive space: 450MB CD-ROM drive Super VGA (800x600) or higher-resolution monitor For connecting with an SPSS Statistics Base Server, a network adapter running the TCP/IP network protocol n Web browser: Internet Explorer 6 or above SPSS Statistics Base 17.0 for MAC OS n Operating system: Apple Mac OS 10.4 (Tiger ) or Mac OS 10.5 (Leopard ) n Hardware PowerPC or Intel processor Memory: 512MB RAM or more; 1GB recommended Minimum free drive space: 800MB CD-ROM drive Super VGA (800x600) or higher-resolution monitor n Web browser: Safari 1.3.1, Mozilla Firefox 1.5, or Netscape 7.2 n Java Standard Edition 5.0 (J2SE 5.0) Additional multithreaded algorithms Additional algorithms have been made multithreaded for improved performance on machines containing multiple processors and multi-core processors. The following procedures are now multithreaded: SORT and Multinomial Logistic Regression in SPSS Statistics Base; and Complex Samples Cox Regression in SPSS Complex Samples. SPSS Statistics Base Server optimizes the multithreading capabilities of these and other SPSS Statistics multithreaded algorithms. While an SPSS Statistics client is limited to four threads, deploying on a server offers an optimized number of threads. This reduces the time needed to prepare and analyze large datasets, which maximizes your organization s IT investment. SPSS Statistics Base 17.0 for Linux n Operating system: any Linux OS that meets the following requirements*: Kernel 2.6.9.42 or higher glibc 2.3.4 or higher Free86-4.0 or higher libstdc++5 n Hardware: Processor: Intel or AMD x86 processor running at 1GHz or higher Memory: 512MB RAM or more; 1GB recommended Minimum free drive space: 450MB CD-ROM drive Super VGA (800x600) or a higher-resolution monitor n Web browser: Konqueror 3.4.1, Firefox 1.0.6, or Netscape 7.2 * Note: SPSS Statistics 17.0 was tested on and is supported only on Red Hat Enterprise Linux 4 Desktop and Debian 4.0 SPSS Statistics add-on modules All SPSS Statistics 17.0 add-on modules require SPSS Statistics Base 17.0. No other system requirements are necessary. 4 Amos 17.0 n Operating system: Windows P or Windows Vista n Hardware: Memory: 256MB RAM minimum 125MB or more available hard-drive space Web browser: Internet Explorer 6
SPSS Statistics Server 17.0 n Operating system: Windows Server 2003 or Windows Server 2008 (32-bit or 64-bit); Sun Solaris (SPARC) 9 and later (64-bit only); IBM AI 5.3 and later; or Red Hat Enterprise Linux ES4 and later; HP-U IIi (64-bit Itanium) n Hardware: Minimum CPU: Two CPUs recommended, running at 1GHz or higher Memory: 256MB RAM per expected concurrent user Minimum free drive space: 300MB Required temporary disk space: Calculate by multiplying 2.5 x number of users x expected size of dataset in megabytes SPSS Statistics Adapter for SPSS Predictive Enterprise Services n Requires SPSS Statistics Base 17.0 and SPSS Predictive Enterprise Services Version comparison chart: new features added to SPSS by version number and by area New feature Version number 17.0 16.0 15.0 14.0 13.0 12.0 11.5 General Switch user interface language SPSS Statistics users on Mac OS and Linux platforms can connect clients to SPSS Statistics Server Desktop versions available on Windows, Mac OS, or Linux Resizable dialogs and drag-and-drop in dialogs Programmability Updated plug-ins for Python,.NET, and R, including support for graphic packages written in R Custom Dialog builder to create user-defined interfaces for existing procedures and user-defined procedures Addition of Python as a front-end cross-platform scripting language Ability to create a data source, including variables and cases, without having to import the active data source into SPSS Control the flow of your syntax jobs or create your own user-defined algorithms using external programming languages (through the SPSS Programmability Extension) Python programming language included on the SPSS CD Ability to create first-class, user-defined procedures Syntax control of output files Call front-end Python scripts or scripting APIs explicitly from within back-end Python programs. Predictive Enterprise Several multithreaded procedures for improved performance and scalability Support for the Predictive Enterprise View, a common data interface that can be defined once and used by all SPSS Inc. analytic tools SPSS Statistics Adapter for SPSS Predictive Enterprise Services (added in SPSS 14.0.1) Updated PMML to include transformations Administrative enhancements in SPSS Statistics Server, including optimized multithreading, virtualization support and a file in use message to reduce errors in data created by more than one person writing to an SPSS file at the same time 5
Version comparison chart: new features added to SPSS by version number and by area New feature Version number 17.0 16.0 15.0 14.0 13.0 12.0 11.5 Single administration utility for SPSS Server, Clementine, and SPSS Predictive Enterprise Services platforms Stripe temporary files over multiple disks for increased performance (in SPSS Statistics Server) Data-free client (in SPSS Statistics Server) Support for Open SSL (in SPSS Statistics Server) In-database data preparation (sort and aggregate) to improve performance (in SPSS Statistics Server) Score data using PMML models created with SPSS, Clementine, and AnswerTree (in SPSS Statistics Server) Predictor Selection and Naïve Bayes algorithms (in SPSS Statistics Server) Data access and data management Read access to SPSS Statistics data files as an ODBC/JDBC data source, allowing these files to be read using SQL Codebook procedure to automatically describe the dataset Improved Data Editor Ability to customize variable view Improved syntax editor, with auto-completion, auto-indentation, color-coding and error-coding of syntax, gutter to display line numbers and break point, and stepping through of syntax jobs Spell-checking of long text strings Spell checking for value labels and variable labels Ability to sort by variable name, type, format, etc. Unicode support Syntax to change string length and basic data type of existing variables Creation of value labels and missing values on strings of any length Ability to set a permanent default working directory Define variable properties tool Date and Time Wizard Export to Database Wizard Identify Duplicate Cases tool Clone dataset command Ability to open multiple datasets within a single SPSS session Export data to recent versions of Excel, including Excel 2007, and SAS Long variable names (up to 64 bytes) Very long text strings (up to 32,767 bytes) Long value labels (up to 120 bytes) Custom Attributes for user-defined meta data in the SPSS Data Editor Read recent SAS files Read/write Stata files Export to Dimensions Data Model OLE DB data access (Windows only) Restructure Data Wizard Visual Binner to easily bin data (for example, break income into bands of $10,000) Optimal Binning (in SPSS Data Preparation add-on module) Subset variable views 6 Features subject to change based on final product release.
Version comparison chart: new features added to SPSS by version number and by area New feature Version number 17.0 16.0 15.0 14.0 13.0 12.0 11.5 Analysis SPSS EZ RFM add-on module enables business users to categorize customers based on the recency, frequency, and monetary value of their purchases SPSS Neural Networks add-on module Multiple imputation of missing data (in SPSS Missing Values add-on module) Complex Samples Cox Regression (in SPSS Complex Samples ) Latent Class Analysis in Amos Partial Least Squares regression** Support for R algorithms** Regularization methods (in SPSS Categories add-on module): Ridge regression, the Lasso, Elastic Net Model selection methods (in SPSS Categories): 632(+), bootstrap, Cross Validation (CV) Multiple correspondence analysis (in SPSS Categories) Preference scaling (in SPSS Categories) Nearest Neighbor analysis, which can be used for prediction or for classification (in SPSS Statistics Base) Median transformation function in COMPUTE procedure Option to use aggressive versus conservative rounding in COMPUTE procedure Create new variables the contain the values of existing variables from preceding or subsequent cases TwoStep cluster analysis (in SPSS Statistics Base) Descriptive ratio statistics Generalized linear models (in SPSS Advanced Statistics add-on module) Generalized estimating equations (in SPSS Advanced Statistics) Ordinal regression to model ordinal outcomes (in SPSS Statistics Base ) SPSS Complex Samples add-on module Complex samples general linear model and logistic regression (in SPSS Complex Samples add-on module) Complex samples ordinal regression (in SPSS Complex Samples) SPSS Decision Trees add-on module Validate Data procedure (in SPSS Data Preparation add-on module) Anomaly Detection for multivariate outliers (in SPSS Data Preparation) Enhanced SPSS Forecasting add-on module with Expert Modeler Bayesian estimation MCMC algorithm (in Amos structural equation modeling software) Data imputation, including multiple imputation (in Amos structural equation modeling software) Estimation and imputation of ordered-categorical and censored data (in Amos structural equation modeling software) Run significance tests on multiple response variables, excluding categories used in subtotal calculations (in SPSS Custom Tables add-on module) Features subject to change based on final product release. **Available at SPSS Developer Central; requires the SPSS Programmability Extension SPSS Decision Trees was formerly called SPSS Classification Trees, SPSS Forecasting was formerly called SPSS Trends, and SPSS Custom Tables was formerly called SPSS Tables. 7
Version comparison chart: new features added to SPSS by version number and by area New feature Version number 17.0 16.0 15.0 14.0 13.0 12.0 11.5 Graphs GraphBoard integration, enabling users of SPSS Statistics products to deploy new or customer graph templates created in the new SPSS Viz Designer stand-alone module Presentation graphics system Chart Builder user interface for graphics Support for SPSS Inc. s Graphics Production Language (GPL) Dual-Y axis and overlay charts Enhanced process control charts 2-D line charts (both axes can be scale axes) and charts for multiple response sets Population pyramids (also called mirror charts or dual charts), 3-D bar charts, and dot charts (also called dot density charts) Additional chart display features/options, including paneled charts and error bars on categorical charts Output Find and Replace feature in the Output Viewer Enhanced SPSS Custom Tables module with table preview builder and inferential statistics Wrapping and shrinking of wide tables in Word and PowerPoint Create a new worksheet in Excel by appending rows and columns Export output to Microsoft Excel and Word Export output to Microsoft PowerPoint Export output to PDF Syntax to automate report production Output Management System (turn pivot table output, such as SPSS data files, ML, and HTML, into data/input) Interactive interface for the output management system Switch output language SmartReader to allow the viewing and pivoting of SPSS Statistics output Licensing improvements Network license reservations and priority settings Network commuter license License manager redundancy Help SPSS Manuals on CD, featuring manuals in PDF format for SPSS Base and all add-on modules Statistical Coach Tutorial Chart tutorial What s This? (context-sensitive help) Features subject to change based on final product release. To learn more, please visit www.spss.com. For SPSS office locations and telephone numbers, go to www.spss.com/worldwide. SPSS is a registered trademark and the other SPSS products named are trademarks of SPSS Inc. All other names are trademarks of their respective owners. 2008 SPSS Inc. All rights reserved. S17CMP-0608