SUDAAN’s success story is unusual, born of necessity, expertise, science and circumstance. Its development is founded upon the previous work of pioneering statisticians like Drs. Gertrude Cox and Daniel Horvitz, who were early staff members at the Research Triangle Institute (RTI). SUDAAN's evolution mirrors the explosive growth of computing technologies since the early 1970s. Its unique niche in statistical software sales today attests to the vision and commitment of its creators at RTI.
SUDAAN has evolved from a single general survey analysis procedure—STDERR—developed in 1970. Its creation was a response to the need for statistical software that supported the analysis of data from complex sample surveys. Although several versatile software packages were available at the time—SPSS, SOUPAC, BMDP, P-stat, OSIRIS, SAS, and Minitab—none was designed to compute standard errors that took into account the survey design. This void became the underlying reason for development of all of RTI’s subsequent statistical analysis software. Soon STDERR became a procedure under SAS and began evolving.
In 1976, RTI committed to create a multi-procedure package, a process that proceeded swiftly with the advice and assistance of the SAS Institute. Four routines were soon implemented as SAS procedures under SAS76 on IBM mainframes. These procedures were used regularly by statisticians at both RTI and the University of North Carolina at Chapel Hill.
As the need for the unique capabilities of this statistical software became more evident, many other sources provided funding for its ongoing development. The National Center for Health Statistics (NCHS) has long been a major supporter; the National Center for Education Statistics (NCES) and the Environmental Protection Agency (EPA) have supported modification and development of the product as well. RTI began licensing these RTI SAS procedures in 1984.
The evolution of the RTI SAS procedures into the SUDAAN product began in the mid- 1980s, with RTI’s vision of software capable of performing high-volume computations on a microcomputer. As development and testing of the product proved successful, RTI began to develop a free-standing package for a DEC VAX environment; its growth was to parallel the development of the MS-DOS version. Eventually, that development included creation of software for DEC VAX mainframes and workstations running under VMS and ULTRIX; for PCs and compatibles under MS-DOS, Windows 3.1, Windows 95, Windows 98, and Windows NT; for Sun SPARC stations under SunOS and Solaris; and for IBM RISC System 6000s under AIX. They met RTI’s goal of complete compatibility among products.
Today RTI continues to expand SUDAAN’s capabilities, staying true to the tradition of developing statistical software that supports the betterment of research and analysis efforts among a variety of users. Although originally introduced for the statistical analysis of sample survey data from stratified, multi-stage cluster samples, the SUDAAN software has proved to apply directly to the analysis of clustered efficacy and safety data from clinical trials, toxicology, and epidemiology studies that are routinely conducted by the pharmaceutical industry. SUDAAN now has the capability to analyze correlated continuous, binary, ordinal, time-to-failure, and count data, in addition to providing descriptive statistics.
After nearly 40 years, the tools have changed greatly, but RTI remains committed to providing state-of-the-art solutions that are effective responses to the evolving needs of SUDAAN users around the globe.
The SUDAAN product you use today is the result of many years of effort by RTI staff and by trusted colleagues outside RTI. Since 1972, when Dr. Babubhai Shah proposed to develop the single survey data analysis procedure that has evolved into today's SUDAAN, these contributors have committed considerable resources to its ongoing development. Here, we acknowledge the vital contributions made by some of the more prominent participants.
Comments, suggestions and contributions from the following individuals outside of RTI have been particularly invaluable to the success of SUDAAN:
Several institutions over the years have provided advice, and funding assistance through research contracts:
Many past and current RTI statisticians and programmers have made valuable contributions, either through their work on the SUDAAN software, or as users of the SUDAAN procedures. We sincerely appreciate all the hard work and dedication each has put into making the SUDAAN product the premier statistical software package that it is today.
In addition, we gratefully acknowledge the past and present officers of RTI who have offered their support for SUDAAN over the years. They shared our belief that the expertise and enthusiasm of RTI staff would result in successful ventures, and invested RTI's resources accordingly. We also thank the current officers for their backing of recent software and language development, and for their assurance to SUDAAN users everywhere of their continued support of the product's development in the foreseeable future.
Finally, we would like to acknowledge the following RTI staff for their important contributions to SUDAAN Release 11:
The success of SUDAAN today is, in large part, due to the dedication and contributions of these individuals. RTI International thanks them for their hard work and perseverance. We look forward to their continued involvement with this very successful and respected program for many more years to come.