Recent data indicate that 27.6% of 8th and 10th graders have used alcohol in the previous month (Johnston et al., 2003). While intrapersonal development is a dynamic that stimulates experimental behavior in adolescents, family processes are also importantly involved. Parental monitoring, open positive communication and parental promotion of independence have been shown to buffer adolescent alcohol use (Barnes, et al, 2000). However, other studies indicate that poor family communication and a lack of monitoring are predictors of adolescent alcohol use (Dishion & McMahon, 1998). The purpose of this study is to classify families of adolescents into latent classes based on family processes in an attempt to answer the following questions: How many classes of families are there? What is the prevalence of classes? What are the nature of the classes? Which items differentiate classes? We conducted a latent class analysis with Mplus Software using the Add Health Data (Wave II). Latent class analysis is a statistical method for identifying unmeasured class membership, which is estimated from the observed variables. The sample consisted of 13,570 adolescents, 6964 females and 6606 males representing varying ethnic and socioeconomic groups. Measures included items which tapped family communication, parental monitoring, family emotional climate and rates of adolescent drinking. Results indicate that there were two classes: class 1 comprised 36.7% and class 2, 63.6% of the sample. Estimated means for the eleven family variables based on posterior probablity of class membership were: Class 1 – (1) .93 (2) 1.11 (3) 1.12 (4) .99 (5) .87 (6) .95 (7) 1.16 (8) .86 (9) 1.00 (10) .88 (11)1.89. Covary with alcohol -.06; Class 2 – (1) 1.90 (2) 2.01 (3) 2.24 (4) 2.32 (5)2.03 (6) 2.15 (7) 2.43 (8) 1.63 (9) 1.87 (10) 1.76 (11) 2.05 Covary with alcohol +.013. All of the items discriminated class membership well. The findings show that families in class 1 are emotionally close, have slightly better communication and have higher levels of parental monitoring; all of which are negatively related to adolescent alcohol use. On the other hand, class 2 families have lower means on the measures which are positively related to alcohol use. Results indicate that family processes can be a risk or a protective mechanism. Adolescence is a difficult time in the family life cycle. These findings suggest the importance of interventions which improve family communication, parental monitoring and cohesion building. References Barnes, G.M., Reifman, A.S., Farrell, M.P. Deatcheff, B.A. (2000). The effects of parenting on the development of adolescent alcohol misuse: A six wave latent growth model. Journal of Marriage and Family, 62, 175-186. Dishion, T.J. & McMahon, R.J. (1998). Parent monitoring and the prevention of child and adolescent problem behavior: A conceptual and clinical formulation. Clinical Child and Family Psychology Review, 1, 61-75. Johnson, L.D., O'Malley, P.M., Bachman, J.G. (2000). Monitoring the Future. Rockvill, MD: National Institute of Drug Abuse.