A student has studied a sample of 150.000 recently released convicts from 550 medium security penitentiaries across all the 50 states in the U.S., in the last year. These convicts had similar income levels and living conditions prior to their incarcerations, some of them were re-incarcerated within a year while some were not. The dependent variable reinc is a binary variable which takes the value 1 if a recently released convict is re-incarcerated within a year, and 0 otheMise. The student wants to measure the probability that a recently released convict is re-incarcerated within a year, Pr[reinc= 1], given the number of prior incarcerations (PI, measured by the number of times a convict was incarcerated prior to the recent release). The researcher estimates the following logit regression model: A Pr[reinc= 1 |FI] = F[- 1.16 + 0.42mi. (0.76) (0.11) Standard errors are given in parentheses. Given that the sample consists of recently released convicts, any random individual from the sample has at least one prior incarceration. The estimated difference in the probability that a single recently released convict will be re-incarcerated within a year, of a change in the number of prior incarcerations from 'I to 2, is % . (Round your answer to two decimal places.)