
Estimating psychometric functions in forcedchoice situations: Significant biases found in threshold and slope estimations when small samples are used
Studied the behavior of curvefitting methods such as maximumlikelihood and probit methods for the special case of forcedchoice experiments, in which the probability of a S making a correct response by chance is not zero. A mathematical investigation of the variance of the threshold and slope estimators shows that the accuracy of the methods is much worse and their sensitivity to the way data are sampled is greater than in the case in which chance level is zero. Further, Monte Carlo simulations show that, in practical situations in which only a finite number of observations are made, the mean threshold and slope estimates are significantly biased. The amount of bias depends on the curvefitting method and on the range of intensity values, but it is always greater in forcedchoice situations than when chance level is zero. (PsycINFO Database Record (c) 2005 APA, all rights reserved)
URL 