GLIMMPSE is an open-source application for calculating power and sample size. GLIMMPSE has been designed so that researchers and scientists with varying levels of statistical training can have access to reliable power and sample size calculations. For optimum usability, GLIMMPSE provides two different modes. In Guided Mode users receive step-by-step guided instructions for entering data in order to obtain power and sample size outputs. In Matrix Mode users receive less guidance, and are assumed to possess in-depth statistical training.By using GLIMMPSE, researchers and scientists can calculate reliable values for power and sample size for designs with normally distributed outcomes. GLIMMPSE supports a variety of multilevel and longitudinal studies. GLIMMPSE can calculate power and sample size for common statistical tests and models including: One sample t-test Paired t-test Two sample t-test Analysis of variance (ANOVA) Analysis of covariance (ANCOVA) Repeated measures analysis of variance Multivariate analysis of variance (MANOVA) Multivariate analysis of covariance (MANCOVA)The web-based version of GLIMMPSE is available at http://glimmpse.samplesizeshop.org.Theoretical details are available at http://www.jstatsoft.org/v54/i10.Development TeamGLIMMPSE was developed by Dr. Sarah Kreidler at the University of Colorado Denver. Dr. Sarah Kreidler led a team of software engineers including Dr. Aarti Munjal and Ms. Uttara Sakhadeo. Dr. Munjal and Ms. Sakhadeo were instrumental in the extension of GLIMMPSE to mobile platforms. Mr. Scott Stodghill and Dr. Wes Munsil began supporting GLIMMPSE in 2015.Statistical support and technical writing were provided by Dr. Keith Muller, Dr. Deb Glueck, Dr. Anna Barn, Mr. Zacc Coker-Dukowitz, and Brandy Ringham.AcknowledgementsGLIMMPSE extends an existing program called POWERLIB. Bercedis Peterson and Keith Muller wrote POWERLIB for SAS PROC MATRIX (Version 0) in 1983. In 1992, Lynette Keyes and Keith Muller converted POWERLIB to SAS/IML (Version 1). Keyes also created a new manual, based on the PROC MATRIX manual. Douglas Taylor wrote code for creating confidence limits for power values based on random estimates in 2001, which Keith Muller then used to enhance POWERLIB. In 2003, Jacqueline Johnson and Keith Muller created Version 2 of POWERLIB (Johnson et al. 2009). Additionally, James Slaughter greatly improved the _PROBF module, and Matthew Gurka conducted extensive tests. At the same time, the related manual was significantly revised from the previous version.GLIMMPSE extends POWERLIB to include the work of Glueck and Muller (2003). Development of GLIMMPSE 1.0.0 was funded by an American Recovery and Re-investment Act supplement (3K07CA088811-06S) for NCI grant K07CA088811. Both the parent grant and supplement were awarded to the University of Colorado Denver (Deb Glueck, PI).Additional funding was provided by NIDCR 1 R01 DE020832-01A1 to the University of Florida (Keith Muller, PI; Deb Glueck, University of Colorado site PI). The funding from NCI allowed the development of the GLIMMPSE software. The additional funding from NIDCR permitted extensive architecture changes which will eventually support power for the general linear mixed model, as well as supporting beta testing and software release activities.Additional support for Dr. Muller was provided in part by NIHNIDCR grants U54-DE019261, NIHNCRR grant K30-RR022258, NIHNHLBI grant R01-HL091005 and NIHNIAAA grant R01-AA013458-01.
What's new in version 2.2.2
Bug fixes:[SSS-120] - Finding sample size often fails, but reports the wrong answer.[SSS-122] - Glimmpse fails to allow time = 0.[SSS-123] - Sample size calculations are incorrect for Geisser-Greenhouse and Huynh-Feldt.[SSS-124] - "No mean difference" calculation should not require matrix entries to be identically zero.