Predicting Effectiveness of Wood Preservatives From Small Sample Field Trials

Carol L. Link, Rodney C. DeGroot

Abstract


Field tests of wood preservatives use groups of stakes treated at various retentions. Although only an average value is reported for a given group of stakes, the lifetime of individual stakes is quite variable. This paper explores presentations of data that reflect such variability. We also consider the feasibility of predicting the effectiveness of a preservative before all stakes fail. For sample sizes of ten replicate stakes, we suggest that reports include box plots of the actual failure times, and that studies use the sample median for the reported lifetime value rather than the sample mean and report the first quartile as a lower bound for the population average.

Keywords


Box plots;censored data;nonparametric statistics

Full Text:

PDF

Refbacks

  • There are currently no refbacks.