Treating bias as variance for experimental design purposes

When an empirical model is fitted to data, bias can arise from terms that have not been incorporated, and this can have an important effect on the choice of an experimental design. Here, the biases are treated as random, and the consequences of this action are explored for the fitting of models of first and second order.     

Laser,photolysis of formic acid vapor at 222 nm. quantum yields of stable products

Quantum yields of the stable products, CO, CO₂ and H₂, formed in the excimer laser photolysis of formcc acid vapor at 222 nm and 296 K, have been determined. The results, taken together with the primary quantum yield of OH and the mechanssm of the subsequent reaction of OH with formic acid, studied earlier in our laboratory, are consistent with the photolytic channels, . Although calculated and measured yields from the photolysis of a range of pressures of formcc acid are in good agreemen,, the mechanism is not unique, and molecular photodecomposition paths may occur for the monomer to some extent.     

Simplicity and credibility: A counterstrategy

A counterstrategy to conventional strategies which increase technical complexity to achieve small gains in efficiency and power is to trade of ‘bestness’ for simplicity. In public issues (e.g., neonatal hypothyroidism at Three Mile Island) simplicity enhances credibility.