Rigid polyisocyanurate-modified polyurethane (PIR) foams remain the insulation materials of choice for many commercial and residential construction applications. PIR foam boards, produced by a continuous lamination process, provide an extremely cost-effective combination of thermal resistivity and structural integrity. Utilizing innovative chemistry and processing, the excellent performance of PIR boardstock has been sustained, while achieving compliance with recent environmental regulations concerning ozone depletion. Similarly, steel faced polyurethane (PUR) foam panels with fire retardant additives have also come to play an important role in the construction industry. In construction applications, PIR boards and PUR panels are exposed to widely disparate temperature extremes. However, the structural performance of these rigid foams is typically measured only at certain discrete temperatures. Compressive properties are normally measured at room temperature, and often only in one dimension. Dimensional stability is commonly assessed only under one cold (usually-20°F) and one hot/humid condition. The results of such testing are often taken to be representative of a foam's performance under all possible exposure conditions; clearly a bold assumption. This paper details a fundamental investigation into the influence of temperature on the compressive properties and dimensional stability of closed cell rigid foam. It is shown that these critical parameters are dependent on the physical and morphological properties of the polymeric cellular structure, as well as the thermodynamic properties of the gases used to expand the foam. The compressive properties of typical PIR and PUR foams were measured under a wide range of temperatures in three orthogonal directions. This data was fitted into a model based on fundamental material parameters, and the relationship between dimensional stability and compressive strength in an anisotropic foam was examined. Foams expanded with chlorofluorocarbons (CFC's), hydrofluorochlorocarbons (HCFC's), and several zero ozone depletion potential (ODP) blowing agents were evaluated, resulting in a methodology for predicting field performance under a wide variety of environmental exposure conditions. Foam cell morphology can have dramatic influence on foam structural performance. As the dimensional stability of rigid foams is controlled by the mechanical properties in the weakest direction, foams of similar matrix composition and density can exhibit tremendously different dimensional stability performances due to variations in cell orientation. The properties of the blowing agent also play a critical role in dimensional stability. While -20°F was an appropriate test temperature for CFC and HCFC blown foams, other temperatures may be more appropriate for the best possible assessment of zero ODP foam dimensional stability. The role of matrix composition is also significant. PUR foams plasticized by fire retardant additives appeared to exhibit stronger responses to temperature than highly crosslinked PIR foams. Ultimately, an effective combination of matrix strength (chemical composition and density), blowing agents, and processing (minimal cell orientation) for the application in question is the key to dimensionally stable rigid foam. The methodology presented validates the selection of a relatively weak matrix (PUR, low density), in conjunction with an HCFC blowing agent and excellent processing (isotropic cells), for metal faced panels. The methodology also supports the stronger matrix (PIR, higher density), necessary for permeably faced HCFC boardstock, due to the inherent cell orientation which results from the continuous lamination process.