PSPP

PSPP is an application for statistical analysis of data. PSPP is a completely free replacement for the commercial application SPSS (now PASW) and is very similar to it. It's syntax files and data are also compatible with SPSS.

PSPP can perform several data transformation (including count, recode, weighting and handling of missing values), compute descriptive statistics (frequencies, descriptive statistics), compute crosstabs , T-tests (independent samples T-test, paired samples T-test and one-sample T-test), One Way ANOVA, linear regression, bivariate correlation, factor analysis, reliability, some non-parametric tests (Chi-square and binominal) and ROC curve. The number of statistical operations it can perform is growing. PSPP was designed to perform these tests as quick as possible, regardless of the number of cases.

Some key features:

• PSPP supports over 1 billion cases and over 1 billion variables;
• syntax (SPS) and data (SAV) files are compatible with SPSS;
• it is cross platform (works on multiple operating systems, including Linux and Windows) and Free Software, licensed under GPLv3 or later;
• it is absolutely free and has no expiration date;
• you can using a command line interface (pspp) a graphical interface psppire);
• it can produce output in text, PostScript, PDF, HTML or OpenOffice.org text format;
• it is interoperating with Gnumeric, OpenOffice.Org and other free software;
• statistical procedures are performed fast, even with very large data sets.

The official PSPP website is located at http://www.gnu.org/software/pspp/, binary packages of PSPP for several platforms cah be downloaded from pspp.kiberpipa.org.

PSPP is available in two branches:

• stable branch, which is stable and thoroughly tested, but does not has some new functions;
• master branch contains the latest development, with more functions, but is not so stable and tested as stable branch yet.

In this wiki you can find detailed instructions on how to use PSPP (graphical interface):

• PSPP overview (windows and menus)
• Working with data files
• Test data sets
• Data analysis
  • Basic descriptive analysis
  • Recoding, computation of new variables and missing values
  • Weighting
  • Selecting, filtering and sampling data
  • Crosstabs
  • Explore
  • Compare means (one sample T-test, independent samples T-test, paired samples T-test and one-way ANOVA)
  • Bivariate correlation
  • Factor analysis (Principal Component Analysis and Principal Axis Factoring)
  • Reliability
  • Linear regression
  • Nonparametric statistics (Chi-square and Binominal)
  • ROC curve
• PSPP to SPSS comparison (are computations done the same way?)
• Video guides (by Douglas A. Ferguson, College of Charleston, USA)
• Other wiki resources - further reading about PSPP, free statistical software and statistical methods
• Building binaries
  • Building PSPP binaries for Mac OS
  • Building PSPP binaries for Ubuntu Linux

Wiki is maintained by Matej Kovačič. You can contact me by e-mail matej dot kovacic at owca dot info.