a distribution in which the variables may take on both a discrete and a continuous range of values. This may occur when a continuously measured variable is abruptly topped off or bottomed out at an arbitrary value for a number of participants in the sample. For example, consider a health researcher studying the effects of a smoking cessation intervention who records the number of days since a participant quit smoking. For those participants who quit, the number of days will reflect an infinite series of potential values, whereas for those who continue to smoke the number of days will be the single finite value of zero.