Table 4.
Regions of significance in models predicting daily consequences from consumption and operationalizations of SAM
| Daily Consequencesa | ||||||||
|---|---|---|---|---|---|---|---|---|
| All βs | ROS βsc, d | ROS Range | p-value in ROSe | |||||
| Variable | Low | High | Low | High | Low | High | Low | High |
| Intercept | −0.115 | −0.035 | - | - | - | - | - | - |
| Level 2 (Person) | ||||||||
| Male | −0.099 | −0.091 | −0.099 | −0.093 | 1 | 10 | .038 | .049 |
| −0.095 | −0.094 | 62 | 151 | .046 | .049 | |||
| Non-white | 0.034 | 0.040 | - | - | - | - | - | - |
| Hispanic/Latino | −0.085 | −0.072 | - | - | - | - | - | - |
| Age | 0.010 | 0.015 | - | - | - | - | - | - |
| School (A) | −0.060 | −0.049 | - | - | - | - | - | - |
| School (B) | 0.122 | 0.135 | 0.122 | 0.135 | 1 | 240 | .017 | .030 |
| AUDIT | 0.014 | 0.015 | 0.014 | 0.015 | 1 | 240 | .001 | .003 |
| CUDIT | 0.004 | 0.004 | - | - | - | - | - | - |
| Level 1 (Day) | ||||||||
| Any daily nicotine | 0.029 | 0.038 | - | - | - | - | - | - |
| Any daily other drugs | 0.077 | 0.093 | - | - | - | - | - | - |
| Non-plant | 0.048 | 0.057 | - | - | - | - | - | - |
| Weekend | 0.088 | 0.094 | 0.088 | 0.094 | 1 | 240 | .001 | .003 |
| SAM (versus CAM) | −0.094 | 0.016 | - | - | - | - | - | - |
| # drinks | 0.049 | 0.076 | 0.049 | 0.076 | 1 | 240 | < .001 | < .001 |
| # drinks * SAM | −0.041 | −0.012 | −0.041 | −0.018 | 3 | 114 | < .001 | .044 |
| # marijuana uses | −0.031 | 0.001 | - | - | - | - | - | - |
| # marijuana uses * SAM | −0.004 | 0.026 | - | - | - | - | - | - |
Note: AUDIT= Alcohol Use Disorders Identification Test; CUDIT-R=Cannabis use Disorder Identification Test Revised. ROS = region of significance. Low and high ROS range represents the range of SAM operationalizations (i.e., the duration between any one drink and one marijuana use on a given day that delineated SAM from CAM) during which a given variable was a significant predictor of daily consequences. For example, ‘37’ falling within the ROS range (from 1 to 240) for a given variable implies that when 37 minutes is used to delineating SAM from CAM, the given variable is significant predictor of daily consequences.
Values for “ROS βs” and “ROS range” are not presented when the predictor is not significant for any operationalization.
Low and high values for “All βs” represent the lowest and highest estimate across all SAM operationalizations.
Low and high “ROS βs” represent the lowest and highest estimate within the ROS.
“All βs” are equivalent to “ROS βs” when “ROS range” is 1–240 (i.e., significant in all operationalizations).
p-values are given within the ROS where significant at p < .05.