We examined attributes of pinon pine (Pinus edulis) associated with the probability of infestation by pinon ips (Ips confusus) in an outbreak in the Coconino National Forest, Arizona. We used data collected from 87 plots, 59 infested and 28 uninfested, and a logistic regression approach to estimate the probability ofinfestation based on plotand tree-level attributes. Pinon pine stand density index was a good predictor of the likelihood of infestation by pinon ips at the plot level, and a cross-validation analysis confirmed that the model correctly classified 82% ofthe cases. Diameter at root collar and pifion dwarf mistletoe infestation level were good predictors of individual tree infestation, and a cross-validation analysis indicated that the model correctly classified 72% of the cases. Results suggest that the occurrence of pinon ips infestations may be related to stress factors associated with increased stocking and pinon dwarf mistletoe infestations.