Data were collected from 645 Latino youth with a mean age of 14.9; 55.3% were female; and 65.6% were from intact families. Most were of Mexican descent (57%) with 18.6% being 1st generation, 73% 2nd generation, and 8% 3rd generation. To account for large percentage of youth who had never used and to model both the probability of having use reported as an Odds Ratio (OR) and the frequency of use among users reported as an Incident Rate Ratio (IRR), we modeled marijuana use as a zero-inflated negative binomial distribution.
PV was significantly associated with an increase in the odds of having used in the past 6 months (OR=1.66), and AM, MM, and FM were all significantly associated with decreases in the odds of having used (OR=.55, .60, and .64, respectively). However, only FM was associated with a decrease in the frequency of use among users (IRR=.75).
Results suggest that youth may turn to marijuana use to cope with the experience of PV (i.e., bullying), and that AM, MM and FM each uniquely buffer against the probability of using. However, once youth have begun use, only fathers seem to be influential in abating that use. This suggests the importance of including fathers in prevention research and suggests different prevention strategies based on what stage of use is being targeted.