PBSTAT : A Web-Based Statistical Analysis Software for Participatory Plant Breeding

Abstract

Indonesian agriculture areas were dominated by variation in agro-ecological and socio-economic conditions implies that formal plant breeding (FPB) programs less effective compare to parcipatory plant breeding (PPB). However to improve genetic gain in PPB, should be supported by simple statistical program. PBSTAT was developed in order to meet the needs of simple statistical software for selection and trials in participatory breeding approach. This software is programmed using PHP scripting language, therefore can be utilized on web platform, and provides easy access for its users to do the statistical analysis. The user can use common spreadsheet software for data entry and web browser to run the analysis. The main feature of PBSTAT is analysis of variance (ANOVA) for analyzing variety trials in one location (using one factor RCBD), multi-locations, and multi-seasons-locations (combined analysis of several RCBD trials across seasons and locations). Mean differences can be compared using Tukey’s HSD method. Other important feature of PBSTAT is stability analysis using Finlay-Wilkinson method. This feature is very useful for the breeder conducting several participatory trials to elucidate which genotypes are stable across environments, and which one are environment-specific.