The management, conservation and theoretical understanding of mixed tropical forests rely heavily on the interpretation of data collected from those forests. Data-derived quantities that are averaged over all species often hide important descriptors of the forest, whereas quantities based on individual species are difficult to interpret due to the diversity of tree species. A compromise between these two analytical approaches is to form groups of tree species according to their characteristics. These groupings are often specific to a particular analysis. An analysis of permanent sample plot data from East Kalimantan (Borneo), Indonesia, has led to the development of a species grouping. The analysis was part of the development of an empirical growth and yield model of mixed tropical forests. The species were grouped based on tree growth and sapling recruitment as represented in the model. The data used with this analytical approach justified the creation of 10 species groups. The four groups consisting of members of the Dipterocarpaceae family, and two groups of fast-growing pioneers (the genus Macaranga and species Anthocephalus chinensis) dominate the classification. The grouping is similar to that used in other analyses in Borneo but is not designed for use solely in modelling and thus is appropriate for many different types of analysis of forest data. The grouping has been applied to the development of a growth and yield model and has been used in descriptions of areas of forest to enable forest management decisions to be made. © 2002 Elsevier Science B.V. All rights reserved.