9.1. History of Statistical Package for the Social Science
In 1965, at Stanford University (San Francisco, USA), two students Nye Norman and Dale Bent specializing in political science, made an effort to find a computer program suitable for statistical data analysis. Soon, they gave up hope in their attempts as existing programs proved useless in part, were poorly constructed, or presented processed information inaccurately. In addition, the use of these computer products varied from program to program.
So, without hesitation, they decided to develop their program with concepts and common syntax. The FORTRAN programming language and the computer of IBM 7090 type were at their disposal. Then, as we know from the history of computer science, the program was a package of punch cards.
One year later (in 1967), the first version of the program for the statistical data analysis was proposed; it could run on IBM 360. By this time, Hedley Hull joined the group of developers.
The original name of the program that the authors gave their product was SPSS, an acronym for Statistical Package for the Social Science.
In 1970 (by that time, sixty product installations had been made), the work on the SPSS program was extended to the University of Chicago, and Norman Nye set up the associated company. The first User’s Guide describes eleven different procedures.
Five years later, SPSS has been already installed six hundred times for different operating systems. From the beginning, serial number versions were assigned to the program. In 1975, the sixth version (SPSS 6) was developed. Later, due to the SPSS 6 version, the 7th, 8th, and 9th version improvements were followed.
So far, 12–26 versions have been developed and are widely used. They are different modifications of the original SPSS. But this variety of modern versions largely mirrors the classical 9.0 and 10.0 versions with a slight improvement in statistical data processing and results displaying in graphic form.
The basic unit of the SPSS program provides access to and management of various data. It contains the most commonly used methods of analysis. Traditionally, two more modules are delivered with SPSS Base (base unit): Advanced Models and Regression Models. These modules cover a number of analytical techniques included in an earlier version of the software for mainframe computers.
SPSS Base is a standard supply. It contains all procedures for input, selection, and correction of data, as well as most of the statistical methods offered in SPSS. Along with simple statistical analysis techniques such as frequency analysis, the calculation of statistical characteristics, contingency tables, correlations, plotting, this module includes many non-parametric tests and sophisticated methods such as multivariate linear regression analysis, discriminant analysis, factor analysis, cluster analysis, analysis of variance, capability study (reliability analysis), and multidimensional scaling.
Due to the expanded scope of application, the name SPSS (Statistical Package for the Social Sciences) was changed to PASW (Predictive Analytics Software) in 2010.