The evaluation of the Woodrock Youth Development Program by LoSciuto and colleagues (1997) used a randomized pretest-posttest control-group design to test the effectiveness of the multi-level program. The cohort design of 367 students (130 in the experimental group, 237 in the control group) was 46.9 percent female and racially diverse (44.4 percent Latino, 19.9 percent White, 11.4 percent African American, 11.2 percent Asian American, 9.3 percent mixed or “other,” and 1.9 percent American Indian).
Participants, who were grade-school students between the ages of 6 and 14, were divided into separate age groups by a median split within the age distribution. Analysis was divided up into two groups: 1) Younger subsample (age 6 through 9), and 2) Older subsample (age 10 through 14).
The evaluation began in four public schools in the Kensington area of Philadelphia, Pa., in November 1994. Questionnaires assessing each of the six outcomes variables and demographic characteristics were administered at the beginning of the school year and again in June 1995.
The Woodrock questionnaire was created to measure students’ attitudes and/or behaviors in the following domains:
- race relations and ethnocentrism
- alcohol, tobacco, and other drug (ATOD) use within the past year
- ATOD use within the last month
- knowledge of harmful effects of drugs and alcohol
- attitudes about the use of drugs and alcohol
Responses were coded so numerical scores were aligned with labels for each scale.
Self-esteem was measured using items from the Hartert Self-Perception Profile for Children. Knowledge and attitude toward ATOD use were measured using several scales from the CSAP Knowledge, Attitude, and Behavior Instrument. Race relations and ethnocentrism were constructed specifically for the evaluation. Changes in the mean scores on the outcome measures were compared from pretest to posttest, controlling for preexisting differences in pretest scores, using the statistical technique of analysis of covariance (ANCOVA).