When the response for an individual unit (well, animal etc) is a binary value, such as alive / dead, or response / no response, the data are treated as quantal.
The responses are assumed to follow a binomial distribution within each dose group. This assumption is required for the calculations of confidence intervals and the p values resulting from statistical tests.
The relationship between the proportion of responders in a dose group and the log dose is modelled using a mathematical formula, representing a sigmoid curve. QuBAS can fit either a logit or a probit model. The parameter values are estimated using the maximum likelihood method, based on the binomial distribution. (Note that any log base can be used for the dose, but the same base must be used when transforming back to obtain calculating the relative potency – see below for details. QuBAS allows a log base of 2, e or 10.)
The contents of Qualitative / dichotomous / quantal response data
