Modern neuropathology has been the recipient of several high-powered quantitative analytical techniques. If properly applied, these techniques provide experimental results with high inter-rater reliability and scientific validity. However, inadvertent methodological problems can be the cause of some of the conflicting results in the schizophrenia research literature. In the present article, we discuss some of the inconsistencies reported in the recent literature and provide suggestions for improving tissue retrieval, selection of anatomical levels, neuronal counts, and neuronometry. We emphasize the importance of shrinkage correction for individual brains and suggest caution in the interpretation of immunocytochemistry and computerized image analysis studies.