How to use this document
This document is intended to introduce researchers to SPSS for the UNIX environment. University Information Technology Services (UITS) at Indiana University has AIX (IBM) Unix operating systems. To learn more about Unix systems you may use Getting Started with UNIX. You may also enroll in an UITS STEPS class by contacting the IT Training & Education. Contact a consultant at the UITS Support Center, or at a UITS Student Technology Center (STC) if you need help. Consultants are on duty at most of the UITS sites. If you need help using SPSS from any UITS computers, contact the UITS Stat/Math Center (e-mail: email@example.com; phone: 812/855-4724 or 317/278-4740).
UITS supports SPSS software under the IBM AIX Libra Cluster. If you want to set up an account on any of the timesharing computers contact the UITS Support Center or visit the webpage:
For more information related to the availability of SPSS at IU, please visit the Availability Web page.
Features of SPSS
SPSS comes with a number of add-on modules along with its Base module. These include the Trends, Tables, and Categories modules. From Release 5 onwards, the Graphics module is incorporated into the Base module. Up to Release 5, the Base module also contains the Statistics module. With Release 5 and above, the Statistics module is separate from the Base module and is divided into Advanced Statistics and Professional Statistics. The Base, Trends, Advanced Statistics, Professional Statistics, Tables, and Graphics modules are available on all central Unix systems. Some features of SPSS are listed below.
Data management capabilities include:
- Detailed labeling of variables and data values; additional documentation of data sets; storage of data and documentation in system files.
- Flexible definition of missing data codes.
- Permanent and temporary transformation of existing variables and computation of new variables; conditional and looping structures for complex data transformations.
- Reading raw data files in a wide variety of formats (e.g., numeric, alphanumeric, binary, dollar, date, and time formats).
- Reading hierarchical and other non-rectangular raw data files.
- Reading, combining, outputting multiple files.
- Reading matrices for input to procedures.
- Flip command to switch the columns and rows in a data set.
- Macro facility to build ones own block of SPSS syntax elements and to control the execution of these blocks.
- Ability to read and write to compressed files.
Statistical procedures for data analysis include:
- The EXAMINE procedure to explore data sets before deciding on the course of data analysis to perform.
- Descriptive statistics, frequency distributions, and cross-tabulations, bar charts, histograms, and scatterplots.
- The RANK procedure, which produces ranks, normal scores, Savage scores, and percentiles for numeric variables.
- T-tests, univariate and multivariate analysis of variance and covariance, including repeated measures and nested designs.
- Multiple regression, nonlinear regression, constrained nonlinear regression.
- Loglinear models for discrete data; probit models.
- Factor and principle components analysis, discriminant analysis, cluster analysis, multidimensional scaling.
- Nonparametric tests.
Besides these capabilities, SPSS add-on modules feature:
- Tables to produce simple or complex tabulation formatted for presentation.
- Trends including time series plots, plots of autocorrelation, partial autocorrelation, cross-correlation function, smoothing, seasonal regression, Box-Jenkins methods, spectral methods and forecasting.
- Categories for doing conjoint analysis and optimal scaling.
When working in a UNIX environment, you often hear about the C-shell (csh), Bourne shell (sh), and Korn shell (ksh). These are simply command language interpreters. They tell the system to act on the command you type in from a terminal. Each shell has some unique features.
For SPSS computing, it makes no difference which shell you use. You access SPSS the same way whether you are in the K-shell or C-shell. Which shell is the default varies according to the system you're using. To change your local login shell, use the chsh command. You can also switch shells by typing ksh (from the C-shell) or csh (from the K-shell). The .login and .cshrc files are executed during login if you use the C-shell; .login and .kshrc files are executed during login if you use the K-shell.
For more on shells, see an introductory guide to UNIX, or The least you need to know about UNIX.
Helpful UNIX commands
Below are a few UNIX commands you may find useful. Italics denote a parameter that you must specify (e.g. filename, directory name, etc.).
ls list files in directory ls -l list files in directory in detail quota display disk quota (if any) history see a list of commands executed so far date print date and time who see a list of all logged in users whoami who is logged on to this account pwd show current directory passwd change password cat file list the contents of the file cat file1 file2 > file3 concatenates file1 and file2 into file more file list file page by page cp file1 file2 copy file1 to file2 mv file1 file2 rename file1 to file2 rm file delete the file head file show the beginning 10 lines of the file tail file show the last 10 lines of the files diff file1 file2 list the file differences wc file count the number of lines, words, and character in the file chmod mode file change the protection mode of the file finger username give information on the user specified. chfn change finger information cd pathname change to directory pathname cd .. move one directory up cd move to the login directory mkdir pathname create a new directory pathname rmdir pathname remove directory pathname man command display UNIX manual entry for command logout end terminal session
Refer to a UNIX commands document for further information.
Editors in UNIX
You may use one of the several editors (e.g., vi, pico, nano, emacs) available from UNIX. Refer to a user's manual or, at the UNIX prompt, type man editor name for online manual. For beginning UNIX users, pico may be the easiest to use. If you're doing e-mail on Shakespeare, you're already using pico, the editor in Pine.
Next: Getting Started
Up: Table of Contents