Table 4.
Measurement models of drug use in the past month
| Model 1 Unstandardized (standardized) | Model 2 Unstandardized (standardized) | Model 3 Unstandardized (standardized) | |
|---|---|---|---|
| Factor loadings | |||
| Number of days alcohol | 1.00 (0.34) | 1.00 (0.39) | 1.00 (0.38) |
| Number of days marijuana | 3.4*** (0.56) | 4.02*** (0.75) | 4.11*** (0.76) |
| Number of days inhalant nitrates (poppers) | 0.11*** (0.24) | 0.05** (0.14) | – |
| Number of days other drugs | 0.89*** (0.49) | 0.63*** (0.40) | 0.62*** (0.39) |
| Any drugs in urine | 0.27*** (0.83) | 0.21*** (0.72) | 0.22*** (0.73) |
| Number of drugs in urine | 0.47*** (1.00) | – | – |
| Variances | |||
| Drug use | 2.02*** (1.00) | 2.63** (1.00) | 2.55** (1.00) |
| Residual variances | |||
| Number of days alcohol | 15.33*** (0.89) | 14.72*** (0.85) | 14.80*** (0.85) |
| Number of days marijuana | 51.45*** (0.69) | 32.56*** (0.43) | 31.77*** (0.42) |
| Number of days inhalant nitrates (poppers) | 0.386*** (0.94) | 0.40*** (0.98) | – |
| Number of days other drugs | 4.90*** (0.76) | 5.44*** (0.84) | 5.52*** (0.85) |
| Any drugs in urine | 0.07*** (0.32) | 0.11*** (0.48) | 0.10*** (0.46) |
| Number of drugs in urinea | 0.00 (0.00) | – | – |
| R2 | |||
| Number of days alcohol | 0.116*** | 0.151** | 0.147*** |
| Number of days marijuana | 0.314*** | 0.566*** | 0.576*** |
| Number of days inhalant nitrates (poppers) | 0.056** | 0.018 | – |
| Number of days other drugs | 0.245*** | 0.162* | 0.150* |
| Any drugs in urine | 0.683*** | 0.523*** | 0.539*** |
| Number of drugs in urine | Undefined | – | – |
| Fit statistics | |||
| χ2 (df) | 120.39*** (10) | 32.70*** (5) | 2.99 (2) |
| CFI | 0.851 | 0.895 | 0.996 |
| RMSEA | 0.136*** | 0.096** | 0.029 |
The model in boldface represents the most theoretically and statistically meaningful measurement model for drug use in the past month; Model 1 includes all six indicators of drug use; Model 2 includes all but the number of drugs in urine indicator; Model 3 includes all but the number of days inhalant nitrates were used and the number of drugs in urine indicators
***
p<0.001,
**
p<0.01,
*
p<0.05
a
Residual variance for number of drugs in urine was negative (and non-significant), resulting in a non-positive definite theta matrix in Model 1; thus it was set to 0