Jump to content

SPSS

From Alfawiki
IBM SPSS Statistics
Company typeSoftware
IndustryStatistical software
PredecessorSPSS Inc.
Founded1968; 58 years ago (1968)
FounderNorman H. Nie, Dale H. Bent, C. Hadlai Hull
FateAcquired by IBM in 2009
Headquarters
Armonk, New York
,
United States
Area served
Worldwide
ProductsStatistical software package including IBM SPSS Statistics, IBM SPSS Modeler, IBM SPSS Amos, IBM SPSS Collaboration and Deployment Services
OwnerIBM
ParentIBM Software Group
Websitewww.ibm.com/products/spss-statistics

SPSS Statistics is a comprehensive statistical software package created by IBM designed for tasks such as data management, sophisticated analytics, multivariate techniques, business intelligence, and forensic investigations. Initially developed by SPSS Inc., the product became part of IBM’s portfolio following its acquisition in 2009. Since 2015, it has been marketed under the name IBM SPSS Statistics.

The acronym originally represented Statistical Package for the Social Sciences, highlighting its early focus on social science research, but was later redefined as Statistical Product and Service Solutions to reflect broader applications.

Overview

SPSS is extensively utilized for statistical analysis across social science disciplines and is also favored by market analysts, healthcare researchers, survey organizations, governmental bodies, educational institutions, various industries, marketing firms, and data mining professionals. The foundational SPSS manual (Nie, Bent & Hull, 1970) played a key role in empowering researchers to independently perform statistical evaluations. Beyond analytics, the software provides capabilities for data handling—including case filtering, restructuring files, and creating derived variables—as well as metadata management through an embedded dictionary within the dataset.

Users can interact with SPSS either through intuitive pull-down menus or via its proprietary fourth-generation command syntax language. Employing syntax facilitates reproducibility of results, streamlines repetitive procedures, and supports complex analytical workflows; certain functions are exclusively accessible via syntax commands. The graphical interface can generate syntax scripts for user modification or review. Jobs may be executed interactively or automated through the Production Job Facility.

A macro scripting language enables the creation of command subroutines. Since SPSS version 14, a Python programmability extension has been incorporated to interact with data dictionaries and dynamically construct syntax commands—superseding the older SAX Basic scripting environment. This extension also allows integration with statistical routines from the free R software package. Starting from version 14, external control of SPSS via Python or VB.NET scripts is possible using dedicated plug-ins; these scripting capabilities have been standard since version 20.

Datasets in SPSS are organized as two-dimensional tables where rows correspond to cases (such as individuals) and columns represent variables (like age). The software supports only numeric and string data types. Data processing proceeds sequentially on a case-by-case basis. File matching operations support one-to-one and one-to-many relationships but not many-to-many joins. A separate Matrix session provides tools for matrix computations and linear algebra.

The user interface offers two distinct views: 'Data View' displays a spreadsheet-like layout of cases and variables without formula functionality; 'Variable View' presents metadata details including variable names, descriptive labels, measurement scales, and other properties. Both views permit manual editing for defining dataset structure or entering data directly—ideal for smaller datasets. Larger datasets are typically imported from external sources such as data entry applications or computer-assisted personal interviewing systems.

SPSS can import and export data from ASCII text files (including hierarchical formats), other statistical software packages, spreadsheets, and databases. It also supports connections to external relational databases through ODBC and SQL interfaces.

Statistical results are stored in a proprietary *.spv file format that supports pivot tables; these outputs can be viewed within SPSS or through a standalone viewer application. Export options include plain text, Microsoft Word documents, PDF files, Excel spreadsheets, HTML pages, XML files, SPSS dataset formats, and various graphic formats such as JPEG, PNG, BMP, and EMF.

Multiple editions of SPSS exist: Gradpacks offer discounted licenses tailored for students; SPSS Statistics Server operates on a client/server model. Additional modules extend core functionality—for instance, the complex samples add-on adjusts analyses for clustered or stratified sampling designs while custom tables facilitates production of publication-quality tables. Licensing arrangements include both annual and monthly subscription plans.

Version 25 was released on August 8, 2017, bringing enhancements like advanced random effects modeling (GENLINMIXED), robust standard error estimation in GLM/UNIANOVA procedures, profile plots featuring error bars, expanded Bayesian statistical methods, and improved charting tools with templates optimized for integration with Microsoft Office applications.

Versions and Ownership Timeline

  • SPSS 1 – 1968
  • SPSS 2 – 1983
  • SPSS 5 – 1993
  • SPSS 6.1 – 1995
  • SPSS 7.5 – 1997
  • SPSS 8 – 1998
  • SPSS 9 – 1999
  • SPSS 10 – 1999
  • SPSS 11 – 2002
  • SPSS 12 – 2004
  • SPSS 13 – 2005
  • SPSS 14 – 2006
  • SPSS 15 – 2006
  • SPSS 16 – 2007
  • SPSS 17 – 2008
  • PASW 17 – 2009
  • PASW 18 – 2009
  • SPSS 19 – 2010
  • SPSS 20 – 2011
  • SPSS 21 – 2012
  • SPSS 22 – 2013
  • SPSS 23 – 2015
  • SPSS 24 – March, 2016
  • SPSS 25 – July, 2017
  • SPSS 26 – 2018
  • SPSS 27 – June, 2019 (27.0.1 released November, 2020)
  • SPSS 28 – May, 2021
  • SPSS 29 – September, 2022
  • SPSS 30 – September, 2024

The initial release of the Statistical Package for the Social Sciences (SPSS) occurred in 1968, created by Norman H. Nie, Dale H. Bent, and C. Hadlai Hull, who established SPSS Inc. as a company in 1975. Early iterations were developed using Fortran for batch processing on mainframe computers such as IBM System/370 and ICL, relying on punched cards for data entry. Starting with version 10 (SPSS-X) in 1983, enhancements included the capability to handle multiple record types.

Prior to version 16.0, separate editions were released for Windows, Mac OS X, and Unix platforms. A notable compatibility issue arose with Mac OS X version 13.0, which did not support Intel-based Macs due to limitations with Rosetta emulation. Additionally, the Windows version of release 15.0 required a patch to function properly on Windows Vista.

Beginning with version 16.0, a single consolidated release was introduced that supported Windows, Mac (offered as a Universal binary), and Linux operating systems via a Java-based graphical user interface.

On July 28, 2009, IBM completed the acquisition of SPSS Inc. at a purchase price of US$1.2 billion. Following a trademark conflict concerning the "SPSS" brand name, the software was temporarily marketed as PASW (Predictive Analytics SoftWare) from late 2009 until early 2010 before reverting to "SPSS: An IBM Company." The full transition was finalized by October 1, 2010, incorporating IBM SPSS into IBM Software Group’s Business Analytics Portfolio alongside products like IBM Algorithmics, IBM Cognos, and IBM OpenPages.

The IBM SPSS suite encompasses additional tools such as IBM SPSS Modeler for data mining and text analytics applications, IBM SPSS Collaboration and Deployment Services enabling real-time credit scoring solutions, and IBM SPSS Amos designed for structural equation modeling.

In a strategic move during 2015, IBM divested its SPSS Data Collection and Dimensions product lines to UNICOM Systems, Inc., which subsequently integrated these offerings into the UNICOM Intelligence portfolio focused on survey design, deployment, data gathering, management, and reporting functionalities.

IDA (Interactive Data Analysis)

IDA (Interactive Data Analysis) originated as a software package developed by the National Opinion Research Center (NORC) at the University of Chicago. Initially deployed on the HP-2000, it later became available on MUSIC/SP under the ownership of SPSS. IDA specialized in regression analysis while providing advanced capabilities for data exploration and statistical modeling.

SCSS - Conversational / Columnar SPSS

SCSS was engineered specifically for IBM mainframe environments and utilized a column-oriented database structure that improved interactive response times. Its core strength aligned with other SPSS products in delivering efficient cross-tabulation analysis to support rapid statistical computations within an online setting.

Project NX

In October 2020, IBM introduced an Early Access Program for the upcoming iteration of SPSS Statistics, internally referred to as Project NX. This advanced version presents a contemporary user interface featuring integrated assistance, tabbed navigation, and improved charting tools, all designed to simplify and accelerate analytical processes.

By December 2021, IBM broadened the program's availability and revealed further enhancements, demonstrating a continued commitment to advancing SPSS in alignment with modern data analysis requirements.

See also

Further reading

  • Argyrous, G. Statistics for Research: With a Guide to SPSS. SAGE, London, 2005.
  • Bryman, A., Cramer, D. Quantitative Data Analysis with IBM SPSS 17, 18 and 19: A Guide for Social Scientists. Routledge, New York, 2011.
  • Levesque, R. SPSS Programming and Data Management: A Guide for SPSS and SAS Users, 4th ed., SPSS Inc., Chicago, 2007.
  • SPSS 15.0 Command Syntax Reference. SPSS Inc., Chicago, 2006.
  • Wellman, B. "Doing It Ourselves: The SPSS Manual as Sociology's Most Influential Recent Book," in Required Reading: Sociology's Most Influential Books, University of Massachusetts Press, Amherst, 1998.

External links