SPSS is IBM’s statistics and data mining toolset, and while it is a very good solution for medium to large organisations, it can be expensive. So here are a couple of free alternatives.
PSPP is a program for statistical analysis of sampled data. It is particularly suited to the analysis and manipulation of very large data sets. In addition to statistical hypothesis tests such as t-tests, analysis of variance and non-parametric tests, PSPP can also perform linear and logistic regression, factor analysis, cluster analysis and much more. It is a very powerful tool for recoding and sorting of data and for calculating descriptive metrics such as skewness, kurtosis and the like.
This alternative has an SPSS look and feel. It comes with terminal and GUI interfaces and supports a variety of data visualisations and data transformation capabilities.
R is certainly the best known and most widely used open source statistics toolkit, but it has a fairly steep learning curve and you certainly need to know what you are doing. Functionality is extended through a large number of libraries which address every conceivable analysis need. There is a wide variety of data visualisation tools and more recent efforts have been targeted at getting R to handle large data volumes. Not for the timid, but for skilled practitioners R is hard to beat.
For those with run-of-the mill analysis needs the free version of Statistician is a Excel add-on. Statistician allows the user to store a data set and perform multiple analyses on it – typically lacking in other products. Summary statistics, correlation and covariance, autocorrelation, sort and rank, sampling, test population mean, test difference in population means, test population proportion, test difference in population proportions, and regression analysis are supported in both Standard and Lite versions. Basic statistical tables are also supported in both, but more advanced features are only available in the Standard Edition. Variance tests, normality tests, non parametric tests, Chi square tests, ANOVA, binary models, count models, multivariate analysis and time series analysis are only supported in the Standard Edition. Quite a lot of features for the low price.