General Statistical Software
Below are brief descriptions of some of the specialized statistical software that we have available in our office. If you need to do a statistical analysis that is not available using a general statistical package, please feel free to contact one of the consultants in the Office of Statistical Consulting. We will be happy to help you select the package below or to find a package elsewhere that best suits your needs.
AMOS computes structural equation models (SEM) and regressions. It automatically uses an iterative process to compute parameter estimates using cases with missing values, and is therefore very useful for regression analyses with missing values. The AMOS web page at Smallwaters has links to many resources about SEM. It is also available from SPSS. There is also an AMOS tutorial available from the University of Texas at Austin's Academic Computing and Instructional Technology Services.
DBMS/COPY translates data files across over 70 different formats, including all popular general statistical packages, many data base programs, and spreadsheets. It is now available with a Cornell site license. For more information, see StatNews #36.
EpiInfo is an easy-to-use database and statistical analysis program that can be downloaded for free from the Center for Disease Control. It allows data to be exported into many different formats.
is a software package for fitting hierarchical multilevel models. A free student version is available.
MIXOR, MIXREG, MIXNO and MIXPREGanalyze two-level mixed models, including logistic and probit regression models for binary or ordered categorical responses. It may be downloaded for free and is available for Windows. For more information, see StatNews #5.
MLwiN is a software package for fitting multilevel models.
nQuery Advisor calculates power estimates and sample sizes for many analyses, including ANOVA and regression.
Stat/Transfer translates data files across 25 different formats, including general statistical packages, many data base programs, and spreadsheets. It is less comprehensive, and less expensive than DBMS/COPY. For more information, see StatNews #36.
StatXact computes exact p-values for non-parametric tests.
SUDAAN is used to analyze data from surveys with complex sampling designs. It can perform descriptive analyses, regressions, and logistic regression. It can be used as either a stand-alone package or as callable software within SAS. In either case, the syntax is similar to SAS's.
WesVar is used to analyze data from surveys with complex sampling designs. It is available free of charge.