The objectives of this study were to estimate genetic parameters and the influence of systematic effects on behavior test results in dogs. Behavior test results on 1,813 Labrador Retrievers (LR) and 2,757 German Shepherd Dogs (GSD) were analyzed. The behavior test included observations on courage, defense drive, prey drive, nerve stability, temperament, cooperation, affability, and gun shyness. Sex and age influenced most of the traits, and seasons of birth and testing and litter size and composition influenced some of the traits. Apart from defense drive in GSD, and courage, nerve stability, hardness, and affability in LR, all traits were heritable, with heritabilities ranging from 0.14 for hardness to 0.38 for affability in GSD, and from 0.03 for affability to 0.56 for gun shyness in LR. Genetic correlations ranged from 1.00 (LR) and 0.95 (GSD) between courage and hardness to -0.01 (LR) and -0.03 (GSD) between gun shyness and defense drive. Most genetic correlations were positive. Correlations with cooperation were mainly negative, especially in GSD. Genetic correlations between courage and defense drive in LR (0.26) and GSD (0.80), between courage and prey drive in LR (0.27) and GSD (0.65), between affability and nerve stability in LR (0.09) and GSD (0.64), between affability and temperament in LR (-0.24) and GSD (0.39), and between cooperation and hardness in LR (0.28) and GSD (-0.67) were significantly different between the breeds. Genetic parameters for defense drive and cooperation in GSD and hardness and gun shyness in LR were genetically different between the sexes. Results of this study indicate that correction for systematic effects is essential when making selection decisions. Estimating breeding values would be a good solution, incorporating both correction for systematic effects and using all genetic links. Genetic parameters need to be estimated for each breed separately.