Table 3.
Impact of SPARC on High-risk Drinking, Alcohol-related Consequences, and Alcohol-related Injuries: Results from 2003–2006 CDS1
| Internal Consistency2 | Least-Squares Means (SE) | Random coefficient model Results |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 (Baseline) | 2004 | 2005 | 2006 | ||||||||
| I | C | I | C | I | C | I | C | Estimate p-value (SE) |
|||
| Drinking | |||||||||||
| Past 30 day use (# days) | N/A | 4.11 (.34) |
4.73 (.34) |
4.18 (.38) |
4.81 (.38) |
4.25 (.31) |
4.88 (.31) |
4.33 (.39) |
4.96 (.39) |
−0.01 (.16) |
0.98 |
| Binge Drinking (# days in past 30 consumed 5 [for males] or 4 [for females] drinks in a row) | N/A | 2.35 (.24) |
2.73 (.24) |
2.37 (.26) |
2.73 (.26) |
2.39 (.24) |
2.73 (.24) |
2.41 (.27) |
2.73 (.27) |
0.02 (.10) |
0.83 |
| Drunk (# days/ typical week) | N/A | 0.55 (.06) |
0.66 (.06) |
0.57 (.07) |
0.66 (.07) |
0.58 (.06) |
0.66 (.06) |
0.60 (.07) |
0.65 (.07) |
0.02 (.02) |
0.25 |
| Most # drinks (past 30 days) | N/A | 4.39 (.27) |
4.84 (.27) |
4.40 (.28) |
4.76 (.28) |
4.40 (.28) |
4.67 (.28) |
4.40 (.26) |
4.58 (.26) |
0.09 (.08) |
0.27 |
| Consequences | |||||||||||
| Moderate Consequences, due to own drinking (Tier I) |
α = 0.86 | 7.78 (.60) |
8.72 (.60) |
8.05 (.64) |
8.87 (.65) |
8.33 (.56) |
9.03 (.56) |
8.60 (.65) |
9.19 (.65) |
0.12 (.26) |
0.65 |
| Severe Consequences, due to own drinking (Tier II) (%1+) |
α = 0.79 | 0.18 (.01) |
0.18 (.01) |
0.17 (.02) |
0.18 (.02) |
0.17 (.01) |
0.18 (.01) |
0.16 (.02) |
0.18 (.02) |
−0.01 (.003) |
0.02 |
| Interpersonal Consequences, due to others’ drinking | α = 0.81 | 1.34 (.12) |
1.08 (.12) |
1.23 (.13) |
1.13 (.13) |
1.18 (.11) |
1.18 (.12) |
1.11 (.13) |
1.22 (.13) |
−0.12 (.08) |
0.16 |
| Community Consequences, due to others’ drinking | α = 0.75 | 5.38 (.38) |
4.49 (.38) |
5.14 (.39) |
4.38 (.39) |
4.90 (.38) |
4.26 (.38) |
4.66 (.39) |
4.15 (.39) |
−0.13 (.16) |
0.41 |
| Alcohol-related Injuries | |||||||||||
| Experienced (%1 +) | αKR-20 = 0.93 | 0.12 (.01) |
0.14 (.01) |
0.11 (.01) |
0.13 (.01) |
0.12 (.01) |
0.11 (.01) |
0.11 (.01) |
0.11 (.01) |
0.01 (.004) |
0.15 |
| Caused to Others (%1+) | αKR-20 = 0.98 | 0.04 (.003) |
0.04 (.003) |
0.03 (.004) |
0.03 (.004) |
0.02 (.003) |
0.03 (.003) |
0.02 (.004) |
0.03 (.004) |
−0.01 (.001) |
0.03 |
Adjusted linear or logistic random coefficient mixed-effects models for CDS data, adjusted for within-school clustering, gender, race, and academic classification. Numbers given are the estimated least-squares means (population margins) or predicted probabilities; (standard error). Estimate (SE) and p-values in last column are from Type III test of condition x time interaction (I vs. C), where the null hypothesis is that there is no difference in the change of the outcome over time between I and C schools; A negative estimate indicates a declining trend over time in I schools compared to C schools.
Internal consistency measures are Cronbach’s alpha (α) or KR-20 (αKR-20), alpha for dichotomous items; N/A = internal consistency measures not available for a single survey item.