Statistics Family

Formerly PASW Advanced Statistics

Enter the realm of powerful and sophisticated analyses

More accurately analyse complex relationships using powerful univariate and multivariate analysis. In addition to the general linear models (GLM) and mixed models procedures, IBM® SPSS® Advanced Statistics now offers the generalized linear models (GENLIN) and generalized estimating equations (GEE) procedures.

IBM® SPSS® Advanced Statistics enables you to:

• Build flexible models using a wealth of model-building options
• Analyse data that exhibit correlation and non-constant variability
• Examine lifetime or duration data to understand terminal events, such as part failure, death, or survival using Kaplan-Meier or Cox regression

Features

Cox Regression
General Linear Modeling (GLM)
            General Factorial
            Multivariate (MANOVA)
            Repeated Measures
            Variance Components
Generalized Linear Models and Generalized Estimating Equations
            Includes Gamma Regression, Poisson Regression and negative binomial
Hierarchical Loglinear Models
Kaplan Meier
Linear Mixed-level Models
            (aka Hierarchical Linear Models)
Survival
Variance Component Estimation

Ensure the stability of your models

Bootstrapping is a useful technique for testing model stability so your analyses generate more accurate results. IBM® SPSS® Bootstrapping makes it simple and easy to do. This module works seamlessly with the Statistics Family by testing a number of analytical procedures, including descriptives, frequencies, means, crosstabs and many more.

IBM® SPSS® Bootstrapping enables you to:

• Create thousands of alternate versions of your dataset, through re-sampling,  for a more accurate view of what is likely to exist in your population
• Increase the accuracy of you analyses by eliminating the outliers and anomalies
• Estimate the standard errors and confidence intervals of a population parameter

Features

PASW Bootstrapping provides the ability to bootstrap a number of analytical procedures including:
Descriptives
Frequencies
Examine
Means
Crosstabs
T-tests
Correlations/Nonparametric Correlations
Partial Correlations
One-way
UniAnova
GLM
Regression
Nominal Regression
Discriminant
Logistic Regression
Binary Multi-nomial Regression
Logistic Ordinal Regression
GENLIN
Linear Mixed Models
Cox Regression

Previously PASW Categories

Predict outcomes and reveal relationships through perceptual maps of categorical data

Unleash the full potential of your data through predictive analysis, statistical learning, perceptual mapping, preference scaling, and dimension reduction techniques, including optimal scaling of your variables. IBM® SPSS® Categories, provides you with all the tools you need to obtain clear insight into complex categorical and numeric data, as well as high-dimensional data.

IBM® SPSS® Categories enables you to:

• Work with and understand nominal and ordinal data with procedures similar to conventional regression, principal components, and canonical correlation
• Use biplots and triplots to represent the relationship between objects (cases), categories, and (sets of) variables in correlation analyses
• Represent similarities between one or two sets of objects as distances in perceptual maps

Features

ANACOR - Correspondence analysis
CATPCA - Principal components analysis for categorical data (replaces PRINCALS)
CATREG - Ridge Regression, Lasso, Elastic Net*
CORRESPONDENCE
HOMALS - Homogeneity analysis
OVERALS - Nonlinear canonical correlation
PROXSCAL - Multidimensional scaling for individual differences scaling with constraints
PREFSCAL - Preference scaling (multidimensional unfolding)
Multiple Correspondence Analysis

Formerly PASW Complex Samples

Correctly and easily compute statistics for complex samples

IBM® SPSS® Complex Samples, provides the specialised planning tools and statistics you need when working with sample survey data. It enables you to make more statistically valid inferences for a population by incorporating the sample design into survey analysis. You can more accurately work with numerical and categorical outcomes and in addition, a new algorithm enables you to predict time to an event.

IBM® SPSS® Complex Samples enables you to use the following types of sample design information:

• Stratified sampling: Increase the precision of your sample or ensure a representative sample from key groups by choosing to sample within subgroups of the survey population
• Clustered sampling: Select clusters, which are groups of sampling units, for your survey
• Multistage sampling: Select an initial or first-stage sample based on groups of elements in your population; then create a second-stage sample by drawing a sub-sample from each selected unit in the first-stage sample. By repeating this option, you can select a higher-stage sample

Features

SamplingWizard/Analysis Plan Wizard
CS Descriptives
CS Tabulate
CS General Linear Models
CS Ordinal Regression
CS Cox Regression (also multithreaded*)
CS Logistic Regression

Formerly PASW Conjoint

Discover what drives your customers’ purchase decisions

IBM® SPSS® Conjoint, gives you a realistic way to measure how individual product attributes affect consumer and citizen preferences. With IBM® SPSS® Conjoint, you can easily measure the tradeoff effect of each product attribute in the context of a set of product attributes; as consumers do when making purchasing decisions. IBM® SPSS® Conjoint can increase your understanding of consumer preferences, enabling you to more effectively design, price, and market successful products.

IBM® SPSS® Conjoint provides answers to critical questions, including:

• Which features or attributes of a product or service drive the purchase decision?
• What market segment is most interested in the product?
• What is the optimal price to charge consumers for a product or service?

Features

CONJOINT - estimate utilities
ORTHOPLAN - for conjoint analysis
PLANCARDS

Formerly PASW Custom Tables

Create the tables you need in next to no time

IBM® SPSS® Custom Tables, enables you to better understand your data and easily report your results to your organisations decision-makers. More than a simple reporting program, IBM® SPSS® Custom Tables provides comprehensive analysis capabilities including the ability to compare means or proportions, select summary statistics and choose from three significance tests.

IBM® SPSS® Custom Tables enables you to:

• Create a table by simply dragging your variables onto the table preview builder
• Export tables to Microsoft® Word or Excel® for use in reports
• Use category management features to exclude specific categories, display missing value cells, add subtotals to your table and more

Features

Inferential statistics
Sig tests on multiple response variables
35 descriptive statistics
Nested Tables
 

Formerly PASW Data Preparation

Improve data preparation for more accurate results

All researchers have to prepare their data prior to analysis. While data preparation tools are included in the IBM® SPSS® Statistics Base product, IBM® SPSS® Data Preparation helps streamline the data preparation process so that you can get ready for analysis faster and reach more accurate conclusions.

IBM® SPSS® Data Preparation enables you to:

• Identify suspicious or invalid cases, variables and data values with validation rules
• More accurately work with algorithms designed for nominal attributes
• Maintain a consistent data cleaning approach from project to project

Features

Validate data
Anomaly detection
Optimal Binning

Formerly PASW Decision Trees

Create classification trees for better identification of groups

IBM® SPSS® Decision Trees creates classification and decision trees directly within IBM® SPSS® Statistics Base to help you better identify groups, discover relationships between groups and predict future events. These highly visual classification trees allow you to present categorical results in an intuitive manner, so you can more clearly explain categorical results to non-technical audiences.

IBM® SPSS® Decision Trees includes four established tree-growing algorithms:

• CHAID: A statistical, multi-way tree algorithm that explores data efficiently and builds segments and profiles with respect to the desired outcome
• Exhaustive CHAID: A modification of CHAID, which examines all possible splits for each predictor
• Classification and Regression Trees (C&RT): A binary tree algorithm which partitions data and produces accurate homogeneous subsets
• QUEST: A statistical algorithm that selects variables without bias and builds accurate binary trees quickly and efficiently

Features

CHAID
C&RT
Exhaustive CHAID
QUEST

Easily identify the right customers

Use IBM® SPSS® Direct Marketing to identify your most valuable customers, whether those are the customers that purchase most frequently, spend the most, or share certain characteristics. Additionally, you are able to determine which customers are likely to respond to certain offers or which customers haven’t purchased anything for a long time and utilise that information to develop marketing plans tailored to cultivate those customer groups.

IBM® SPSS® Direct Marketing enables you to:

• Rank customers by value with RFM analyses
• Segment your customers/constituents into clusters allowing easier customisation of umbrella campaigns to similar customer groups
• Test new campaigns against existing campaigns using control package tests so that the most effective campaigns are selected
• Determine which customers are most likely to respond to an offer using the propensity to purchase tool
• Select the best locations for new stores or agencies using the postal code response tool that identifies a list of postal codes that have shown the highest rate of response to campaigns

Features

RFM analysis
Cluster Analysis
Contact Profiling
Control Package Test
Propensity to Purchase analysis

Formerly PASW Exact Tests

Reach accurate conclusions with small samples or rare occurrences

IBM® SPSS® Exact Tests is used to determine whether a relationship exists between variables when you have a small number of case variables with a high percentage of responses in one category, or when you subset your data into fine breakdowns. IBM® SPSS® Exact Tests provides you with more than 30 exact tests, which cover the entire spectrum of nonparametric and categorical data problems for small or large datasets.

IBM® SPSS® Exact Tests enable you to:

• Extend your analysis by “slicing and dicing” your data into fine subgroups
• Run tests including one-sample, two-sample and K-sample tests on independent or related samples, goodness-of-fit tests, tests of independence in RxC

Features

30 Tests for nonparametric & categorical data

Formerly PASW Forecasting

Build expert time-series forecasts in a flash

Reliable forecasts can have a major impact on your organisation’s ability to develop and implement successful strategies. With IBM® SPSS® Forecasting, you have what you need to predict trends and develop forecasts quickly and easily. Unlike spreadsheet programs, IBM® SPSS® Forecasting has the advanced statistical techniques you need in order to work with time-series data. Regardless of your level of experience, you can analyse historical data, predict trends faster, and deliver information in ways that your organisation’s decision makers can understand and use.

IBM® SPSS® Forecasting enables you to:

• Automatically determine the best-fitting ARIMA or exponential smoothing model, parameters, and predictors
• Model hundreds of different time series at once, rather than having to run the procedure for one variable at a time
• Apply saved models to “what-if” scenarios to optimise your decisions

Features

Expert Modeler
Forecast multiple series (outcomes) at once
Auto Regressive Integrated Moving Average
Autoregression
Exponential Smoothing Methods
Seasonal Decomposition
Spectral Analysis

Formerly PASW Missing Values

Build better models when you fill in the blanks

Missing data can seriously affect your results. If you ignore missing data or assume that excluding missing data is sufficient, you risk reaching invalid or insignificant results. IBM® SPSS® Missing Values uses multiple imputation to replace missing data. It automatically scans the data to determine the best imputation method. You can then specify which variables to impute and the constraints to place on their values.

IBM® SPSS® Missing Values helps you to:

• Quickly detect a serious missing data problem using the overall summary of missing values report
• Determine if missing values for one variable are related to missing values of another with the percent mismatch of patterns table
• Discover significant differences between respondents and non-reposndents using the flexible separate variance t test and cross-tabulation of categorical variables tables

Features

Data Patterns Table
Impution with Means Estimation or Regression
Listwise and Pairwise Statistics
Missing Patterns Table
Multiple imputation of missing data

Formerly PASW Neural Networks

Discover complex relationships in your data

IBM® SPSS® Neural Networks offers non-linear data modelling procedures that assist in the development of more accurate and effective predictive models. This module allows you to gain insight into relationships that would be difficult to discover using traditional, linear statistical techniques.

IBM® SPSS® Neural Networks assists in the following areas:

• Market research: Create customer profiles, discover customer preferences.
• Database marketing: Segment your customer base, optimise campaigns
• Financial analysis: Analyse applicants’ creditworthiness, detect possible fraud
• Operational analysis: Manage cash flow, improve logistics planning
• Healthcare: Forecast treatment costs, perform medical outcomes analysis

Features

Multilayer Perception
Radial Basis Function

Formerly PASW Regression

Make better predictions using regression procedures

IBM® SPSS® Regression gives you an even wider range of statistics so you can get the most accurate response for specific data types. Apply more sophisticated models to your data using its wide range of nonlinear regression models.

IBM® SPSS® Regression includes these procedures:

• Multinomial logistic regression: Predict categorical outcomes with more than two categories
• Binary logistic regression: Easily classify your data into two groups
• Nonlinear regression and constrained nonlinear regression: Estimate parameters of nonlinear models
• Probit analysis: Evaluate the value of stimuli using a logit or probit transformation of the proportion responding

Features

Binary Logistic Regression
Logit Response Models
Multinomial Logistic Regression (also multithreaded*)
Nonlinear Regression
Probit Response Analysis
Two Stage Least Squares
Weighted Least Squares

Formerly PASW Statistics Server

Maximise productivity with IBM® SPSS® Statistics Server

IBM® SPSS® Statistics Server can help your organisation deliver enterprise-strength scalability and enhanced performance. You’ll benefit from added speed, security, scalability, data centralisation and additional procedures.

IBM® SPSS® Statistics Server enables you to do the following:

• Analyse massive data files faster
• Increase productivity
• Use capital resources more efficiently

Features

Naïve Bayes algorithm
Predictor Selection algorithm
Copy-free data access in SQL DBMS

Formerly PASW Viz designer

Easily create and share customised visualisations

A compelling graphic can bring your most highly detailed research findings into sharp focus so people quickly grasp the meaning of your analysis, however creating just the right graphic can sometimes be a challenge. IBM® SPSS® Viz Designer enables you to conceive, create, and share compelling visualisations. From basic, static charts to advanced visualisations that can even include interactive features, you can now easily produce visualisations that make data easier to understand.

IBM® SPSS® Viz Designer enables you to:

• Use your proprietary styles to customise and brand your data visualisations
• Use a a highly intuitive drop-and-drag interface that eliminates the need for advanced programming skills
• Create reusable chart/graph templates so you can share your visualisations with a broad array of end-users

Features

Includes support for the vizML™ visualization markup language
Support for the vizML™ visualization markup language
Datasource support (delimiter-separated, PASW Statistics data files, and relational databases including DB2®, SQL Server™, Oracle®, and Sybase®)

Take your analysis to the next level

Amos® builds models that more realistically reflect complex relationships because any numeric variable, whether observed (such as non-experimental data from a survey) or latent (such as satisfaction and loyalty) can be used to predict any other numeric variable. You can gain additional insight into the causal nature and strength of the relationships among variables.

Amos® enables you to:

• Easily perform structural equation modelling (SEM)
• Quickly create models to test hypotheses and confirm relationships among observed and latent variables
• Move beyond regression to gain additional insight

Features

Confirmatory factor analysis
Structural equation modeling/Path analysis
Multiple imputation of data
Bayesian estimation
Estimation of categorical and censored data
Latent Class Analysis