Doctors' decision processes in a drug-prescription task:: The validity of rating scales and think-aloud reports

The validity of rating scales and of think-aloud protocols were compared in a medical decision task (treatment of high cholesterol values). Twenty doctors were exposed to six case vignettes. All participants were asked to think aloud and 10 of them were also asked to rate their inclination to drug prescription during successive phases of the decision process. Both think-aloud data and ratings indicated an increasing differentiation over time in the inclination to a certain decision between those who finally decided for and against drug prescription, respectively. Think-aloud data seemed to be more sensitive to the directionality of the decision process. The addition of a rating task did not affect the directionality in the think-aloud data. However, rating participants talked more about pharmacological treatment and less about other possible actions. It was concluded that the results generally supported the validity of both think-aloud data and of rating scales as descriptors of decision processes. (C) 2002 Elsevier Science (USA). All rights reserved.