-13-
language, with approximately 8% choosing to complete the study questionnaires in Spanish
rather than English. About 14% reported that they spoke mostly Spanish or only Spanish with
their friends, and 31% reported that they spoke mostly or only Spanish with their families. The
respondents attended schools where Latinos were usually in the majority (27 of 35 schools),
often as members of super-majorities where more than 70% of the students were Latino (19 of
35 schools).
Program Participation. Over 91% of the Mexican/Mexican American students in the 25
intervention schools reported seeing at least one keepin' it REAL video and over 30% reported
seeing all five; 62% reported seeing the PSAs at least once (versus 31% of the control students);
and 68% reported attending a booster activity during the period that they were offered.
Mean Changes Over Time. Table 1 presents the mean change from the pre-intervention
questionnaire to the 14-month follow-up for control students, intervention students and students
in each separate intervention condition (Latino, Non-Latino, and Multicultural version). In
contrast to the change scores that were used in the regression analyses discussed in the next
section, the means in Table 1 were not adjusted for possible pre-intervention differences among
students in the schools assigned to each intervention condition or to the control group. The table
indicates that substance use increased over time for students in all conditions. Reported alcohol
and marijuana use increased the most, with much smaller increases in cigarette use. Compared
to controls, students in each intervention version reported smaller increases in use of all three
substances and also tended to report relatively more desirable outcomes on all other outcomes.
Intervention students reported an increase in refusal confidence whereas control students
reported less confidence in their ability to refuse substance use offers. In the aggregate, students
from all conditions reported stronger intentions or likelihood of accepting substance use offers
over time, yet control students reported larger increases than intervention students. Additionally,