Bioaerosol sampling from various building sites, some of which were subjected to water damage and microbial growth, provided the opportunity to evaluate current recommendations for interpreting bioaerosol sampling data. Data from the ambient environment, a control building, and areas known to have microbial contamination were used as source data for random simulations. The simulations generated two comparison zones from microbial data from the same environment as a test model to identify the failure rate for Spearman's rank correlation. The results of the simulation indicated a failure rate approaching 60%, depending on the number of samples assigned to each zone by the simulation. The simulations indicated that nonparametric statistical treatment of bioaerosol data as currently recommended for building assessment purposes has limitations. An inordinately high Type II error (failure to reject a null hypothesis which is actually not true) is especially apparent when there are small numbers of samples. For example, in applying this methodology to clearance air sampling, a work zone subjected to removal of all moldy materials and a thorough particulate cleaning would still have a significant chance of failure solely due to the variability of the data, if individual samples are evaluated to identify "localized" contamination. This is significant with regard to the number of samples collected and the interpretation of individual samples in rendering evaluations of microbial contamination.

• fungi: assessment, sampling and analysis
• fungi: moisture damaged house -- indoor microbial concentrations
• mathematics: statistics : statistical test, type II error (false alarm)