Instrument
TADPOLE agencies utilize the ADPRC-TADPOLE Substance Attitude and Behavior Survey (SABS 3-5) and (SABS 6-12) instruments to assess the effectiveness of alcohol and drug prevention services. The SABS (3-5) and (6-12) are easy to read, self-response assessment instruments which can be completed within thirty (30) minutes. The SABS (3-5) and (6-12) were developed and tested by ADPRC in 2003.

The SABS (3-5) Assessment Instrument has five (5) subscales:

• Self Esteem – assesses the respondent’s feelings of self-worth. It has a Cronbach’s alpha reliability coefficient of 0.72.
• Drug Attitude – measures the respondent’s attitude toward drug use. The scale has a Cronbach’s alpha reliability coefficient of 0.68.
• Decision Making – measures the respondent’s confidence in making decisions. It has a Cronbach’s alpha reliability coefficient of 0.66.
• Conflict Resolution – assesses the respondent’s ability to resolve conflict without violence, and has a Cronbach’s alpha reliability coefficient of 0.76.
• Family Management – measures parental involvement in the child’s activities and behaviors pertinent to the prevention project. The scale has a Cronbach’s alpha reliability coefficient of 0.62.

The SABS (6-12) Assessment Instrument has five (5) subscales:

• Self Esteem – assesses the respondent’s feelings of self-worth. It has a Cronbach’s alpha reliability coefficient of 0.72.
• Drug Attitude – measures the respondent’s attitude toward drug use. The scale has a Cronbach’s alpha reliability coefficient of 0.77.
• Decision Making – measures the respondent’s confidence in making decisions. It has a Cronbach’s alpha reliability coefficient of 0.67. Conflict Resolution – assesses the respondent’s ability to resolve conflict without violence, and has a Cronbach’s alpha reliability coefficient of 0.47.
• Family Management – measures parental involvement in the child’s activities and behaviors pertinent to the prevention project. The scale has a Cronbach’s alpha reliability coefficient of 0.59.

The ADPRC instrument is a modified version of the Youth Risk Behavior Survey and the Flinders Adolescent Decision Making Questionnaire (Mann, Harmoni, & Power, 1989). It has a range of 0-100. The scores of all items have been adjusted so that the higher score indicates a positive response. The median score is 50, with a score of 50 or greater indicating average/above average performance on the identified subscales.

Data Analysis
Prior to any data analysis, the statistical analyst re-verifies the survey count. Instruments are re-matched by computer using SAS Proc SQL by Participant Identification (ID) Number, Date of Birth, Race, Grade, Gender, and Agency Name.

Differences on matched data between the pretest and posttest on Attitude Scales are computed using a paired t-test design. P-values of .05 or less are used for statistical significance.

In addition, frequencies, medians, and ranges are computed for the demographic variables. Pretest values from the matched data sets are used for demographic variables.