Two alternative versions of the Team Climate Inventory (TCI, Anderson & West, 1994), consisting of either four or five factors, have been introduced by innovation researchers. This study compared the psychometric properties of these two versions by using data obtained from a Finnish sample of 2,265 local government employees. Exploratory factor analysis of the entire sample reproduced both the four- and five-factor versions with good internal consistencies. When a distinction was made between samples with low and high job complexity, significant differences between the four- and five-factor versions emerged. Exploratory factor analysis of the sample with low job complexity suggested that both factor versions obtained clear and interpretable structures. However, only the five-factor version obtained a clear factor structure in the sample of high job complexity. Further LISREL confirmatory factor analyses of both samples suggested that the five-factor structure provided significantly better fit to the data than the four-factor version. In conclusion, the five-factor version of the TCI, including subscales of vision, participative safety, task orientation, support for innovation, and interaction frequency, is preferred to the four-factor version because it is more structurally intact and less likely to be affected by job complexity.