Estimation of Error and Bias in Bayesian Monte Carlo Decision Analysis Using the Bootstrap

Bayesian Monte Carlo (BMC) decision analysis adopts a sampling procedure to estimate likelihoods and distributions of outcomes, and then uses that information to calculate the expected performance of alternative strategies, the value of information, and the value of including uncertainty. These decision analysis outputs are therefore subject to sample error. The standard error of each estimate and its bias, if any, can be estimated by the bootstrap procedure. The bootstrap operates by resampling (with replacement) from the original BMC sample, and redoing the decision analysis. Repeating this procedure yields a distribution of decision analysis outputs. The bootstrap approach to estimating the effect of sample error upon BMC analysis is illustrated with a simple value-of-information calculation along with an analysis of a proposed control structure for Lake Erie. The examples show that the outputs of BMC decision analysis can have high levels of sample error and bias.     

Paternity Establishment for Men’s Nonmarital Births

This paper uses the 2002 National Survey of Family Growth to examine paternity establishment among men’s nonmarital births. Using births as the unit of analysis, I find that paternity establishment for first births (n = 661) is linked to race/ethnicity and relationship status at birth, and these characteristics are associated differently with the timing and location of paternity establishment (in-hospital or at some later point). For higher-parity births (n = 429), paternity establishment for a particular birth is strongly related to prior paternity and fertility behaviors. Paternity is less likely to be established for a higher-parity birth if the father failed to establish paternity for at least one earlier birth, and third or higher-parity births are far more likely to have paternity established at a subsequent point than at the hospital.     

Toward a History of Psychological Expertise

In honor of the 60th anniversary of SPSSI's founding, this issue examines the ascendence of psychological expertise in American society. After World War II, psychology grew as the public accepted the illuminating power and social benefits of psychologists' expert knowledge. In that period, four problems confronted psychological experts as their numbers and influence grew. First, they needed ideas and methods that were new and superior to common sense. Second, experts needed to appear both relevant and objective. Third, psychologists needed to forge alliances with those who held social power in the settings in which they wished to operate. Finally, experts needed to balance the roles of social critic and social engineer.     

The Criterial Crisis of the Academic World*

The author analyzes the crisis in standards in the academic world where symptoms of academic fraud and diminished quality are linked to increasing competitive pressure, demographic forces, and defects in the social system of professional control, examples of a criterial system with mechanical assessment technology that generates paradoxical outcomes. The paper addresses such issues as redundancy of publications and defects in the peer review system, drawing examples from the fields of medicine, science and sociology, It is argued that a decline in academic standards is facilitated by the confusion involved in differentiating criteria from standards. Further analysis of criterial systems is suggested, along with the need for an explicit examination of academic values and quality.