--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- name: log: C:\Users\Owner\Desktop\AggIll\ab_upps_als.smcl log type: smcl opened on: 23 Nov 2012, 12:05:19 . do "C:\Users\Owner\AppData\Local\Temp\STD08000000.tmp" . . * This imputes missing data for the UPPS-P & Affect Lability. . * There was no missing in the raw dataset; however, it's . * standard to do this in our lab (usually due to recoding . * that is already written into the code for various surveys). . foreach var of varlist upps1-upps20 { 2. impute `var' upps1-upps20, gen(i_`var') 3. gen Ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . foreach var of varlist upps21-upps40 { 2. impute `var' upps21-upps40, gen(i_`var') 3. gen Ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . foreach var of varlist upps41-upps59 { 2. impute `var' upps41-upps59, gen(i_`var') 3. gen Ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . . foreach var of varlist als1-als18 { 2. impute `var' als1-als18, gen(i_`var') 3. gen Ri_`var' = round(i_`var') 4. drop i_`var' 5. } 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed 0.00% (0) observations imputed . . * This creates new items from the UPPS-P for recoding . foreach var of varlist Ri_upps2 Ri_upps7 Ri_upps12 Ri_upps17 Ri_upps22 /// > Ri_upps29 Ri_upps34 Ri_upps39 Ri_upps44 Ri_upps50 Ri_upps58 Ri_upps9 /// > Ri_upps47 Ri_upps3 Ri_upps8 Ri_upps13 Ri_upps18 Ri_upps23 Ri_upps26 /// > Ri_upps31 Ri_upps36 Ri_upps41 Ri_upps46 Ri_upps51 Ri_upps56 Ri_upps5 /// > Ri_upps10 Ri_upps15 Ri_upps20 Ri_upps25 Ri_upps30 Ri_upps35 Ri_upps40 /// > Ri_upps45 Ri_upps49 Ri_upps52 Ri_upps54 Ri_upps57 Ri_upps59 { 2. gen Rev_`var' = `var' 3. } . . * This recodes the items created above . foreach var of varlist Rev_Ri_upps2- Rev_Ri_upps59 { 2. recode `var' 1=4 2=3 3=2 4=1 3. } (Rev_Ri_upps2: 481 changes made) (Rev_Ri_upps7: 481 changes made) (Rev_Ri_upps12: 481 changes made) (Rev_Ri_upps17: 481 changes made) (Rev_Ri_upps22: 481 changes made) (Rev_Ri_upps29: 481 changes made) (Rev_Ri_upps34: 481 changes made) (Rev_Ri_upps39: 481 changes made) (Rev_Ri_upps44: 481 changes made) (Rev_Ri_upps50: 481 changes made) (Rev_Ri_upps58: 481 changes made) (Rev_Ri_upps9: 481 changes made) (Rev_Ri_upps47: 481 changes made) (Rev_Ri_upps3: 481 changes made) (Rev_Ri_upps8: 481 changes made) (Rev_Ri_upps13: 481 changes made) (Rev_Ri_upps18: 481 changes made) (Rev_Ri_upps23: 481 changes made) (Rev_Ri_upps26: 481 changes made) (Rev_Ri_upps31: 481 changes made) (Rev_Ri_upps36: 481 changes made) (Rev_Ri_upps41: 481 changes made) (Rev_Ri_upps46: 481 changes made) (Rev_Ri_upps51: 481 changes made) (Rev_Ri_upps56: 481 changes made) (Rev_Ri_upps5: 481 changes made) (Rev_Ri_upps10: 481 changes made) (Rev_Ri_upps15: 481 changes made) (Rev_Ri_upps20: 481 changes made) (Rev_Ri_upps25: 481 changes made) (Rev_Ri_upps30: 481 changes made) (Rev_Ri_upps35: 481 changes made) (Rev_Ri_upps40: 481 changes made) (Rev_Ri_upps45: 481 changes made) (Rev_Ri_upps49: 481 changes made) (Rev_Ri_upps52: 481 changes made) (Rev_Ri_upps54: 481 changes made) (Rev_Ri_upps57: 481 changes made) (Rev_Ri_upps59: 481 changes made) . . * Generation of UPPS-P total scores and subscale alphas . egen uppsmiss = rowmiss(upps1-upps59) . egen alsmiss = rowmiss(als1-als18) . . egen negurgtot = rowtotal(Rev_Ri_upps2-Rev_Ri_upps58 Ri_upps53) if uppsmiss<30 . alpha Rev_Ri_upps2-Rev_Ri_upps58 Ri_upps53 Test scale = mean(unstandardized items) Reversed item: Ri_upps53 Average interitem covariance: .4067499 Number of items in the scale: 12 Scale reliability coefficient: 0.9029 . . egen premedtot = rowtotal(Ri_upps1 Ri_upps6 Ri_upps11 Ri_upps16 Ri_upps21 /// > Ri_upps28 Ri_upps33 Ri_upps38 Ri_upps43 Ri_upps48 Ri_upps55) if uppsmiss<30 . alpha Ri_upps1 Ri_upps6 Ri_upps11 Ri_upps16 Ri_upps21 /// > Ri_upps28 Ri_upps33 Ri_upps38 Ri_upps43 Ri_upps48 Ri_upps55 Test scale = mean(unstandardized items) Reversed item: Ri_upps55 Average interitem covariance: .2607464 Number of items in the scale: 11 Scale reliability coefficient: 0.8599 . . egen persevtot = rowtotal(Ri_upps4 Rev_Ri_upps9 Ri_upps14 Ri_upps19 Ri_upps24 /// > Ri_upps27 Ri_upps32 Ri_upps37 Ri_upps42 Rev_Ri_upps47) if uppsmiss<30 . alpha Ri_upps4 Rev_Ri_upps9 Ri_upps14 Ri_upps19 Ri_upps24 /// > Ri_upps27 Ri_upps32 Ri_upps37 Ri_upps42 Rev_Ri_upps47 Test scale = mean(unstandardized items) Average interitem covariance: .28792 Number of items in the scale: 10 Scale reliability coefficient: 0.8596 . . egen senseektot = rowtotal( Rev_Ri_upps3 - Rev_Ri_upps56) if uppsmiss<30 . alpha Rev_Ri_upps3 - Rev_Ri_upps56 Test scale = mean(unstandardized items) Average interitem covariance: .2655176 Number of items in the scale: 12 Scale reliability coefficient: 0.8073 . . egen posurgtot = rowtotal( Rev_Ri_upps5 - Rev_Ri_upps59) if uppsmiss<30 . alpha Rev_Ri_upps5 - Rev_Ri_upps59 Test scale = mean(unstandardized items) Reversed items: Rev_Ri_upps49 Rev_Ri_upps52 Rev_Ri_upps54 Average interitem covariance: .3693561 Number of items in the scale: 14 Scale reliability coefficient: 0.9120 . . * Generation of Affect Lability total score & alpha . egen ALS = rowtotal(Ri_als1 - Ri_als18) . alpha Ri_als1 - Ri_als18 Test scale = mean(unstandardized items) Average interitem covariance: .5019944 Number of items in the scale: 18 Scale reliability coefficient: 0.9440 . . * Creation of physical aggression count variable from CARE items . egen phys_agg = rowtotal(care1 care6 care7 care8) . . * Creation of logistic variable from physical aggression variable . gen PAlog = 0 if phys_agg == 0 (150 missing values generated) . replace PAlog = 1 if phys_agg>0 & phys_agg~=. (150 real changes made) . . * Creation of count variable of physical aggression among . * individuals who endorsed at least 1 aggressive act . gen newPA = phys_agg if phys_agg>0 & phys_agg~=. (331 missing values generated) . . * Centering of variables in the logistic model . mcenter newPA sex age ALS negurgtot premedtot persevtot senseektot posurgtot note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_newPA | 150 -2.03e-08 5.632476 -4.326667 29.67333 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_sex | 150 -8.74e-09 .4988129 -.5533333 .4466667 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_age | 150 -5.56e-09 1.859302 -1.626667 10.37333 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_ALS | 150 6.61e-08 11.89011 -18.96667 32.03333 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_negurgtot | 150 7.67e-08 6.372919 -11.7 16.3 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_premedtot | 150 -5.68e-08 6.295569 -12.82667 16.17333 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_persevtot | 150 3.76e-08 6.304506 -10.64667 19.35333 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_senseektot | 150 -3.93e-08 6.981228 -20.69333 13.30667 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_posurgtot | 150 7.55e-08 6.720698 -11.2 18.8 . . * Centering of variables in the count model . mcenter PAlog sex age ALS negurgtot premedtot persevtot senseektot posurgtot note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_PAlog | 481 2.48e-09 .4637314 -.3118503 .6881497 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_sex1 | 481 -7.93e-09 .4827589 -.6320167 .3679834 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_age1 | 481 -2.33e-08 2.06811 -1.920998 10.079 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_ALS1 | 481 -8.14e-08 13.12604 -17.20582 36.79418 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_negurgtot1 | 481 -5.08e-09 6.792306 -14.06445 17.93555 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_premedtot1 | 481 -4.71e-08 5.407729 -12.01455 16.98545 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_persevtot1 | 481 -2.58e-08 5.787599 -9.507277 20.49272 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_senseekt~1 | 481 6.81e-08 6.881852 -20.97713 15.02287 note: label truncated to 80 characters Variable | Obs Mean Std. Dev. Min Max -------------+-------------------------------------------------------- C_posurgtot1 | 481 4.80e-09 7.450373 -12.06653 25.93347 . . * Generating interactions among UPPS-P and Affect Lability . * for both the count and logistic models . gen NUxALS1 = C_negurgtot1* C_ALS1 . gen NUxALS = C_negurgtot* C_ALS (331 missing values generated) . gen PRExALS1 = C_premedtot1* C_ALS1 . gen PRExALS = C_premedtot* C_ALS (331 missing values generated) . gen PERxALS1 = C_persevtot1* C_ALS1 . gen PERxALS = C_persevtot* C_ALS (331 missing values generated) . gen SSxALS1 = C_senseektot1* C_ALS1 . gen SSxALS = C_senseektot* C_ALS (331 missing values generated) . gen PUxALS1 = C_posurgtot1* C_ALS1 . gen PUxALS = C_posurgtot* C_ALS (331 missing values generated) . . *Summary statistics & pairwise correlations for all study variables . sum sex age ALS negurgtot premedtot persevtot senseektot posurgtot phys_agg, det Are you male or female? ------------------------------------------------------------- Percentiles Smallest 1% 1 1 5% 1 1 10% 1 1 Obs 481 25% 1 1 Sum of Wgt. 481 50% 2 Mean 1.632017 Largest Std. Dev. .4827589 75% 2 2 90% 2 2 Variance .2330561 95% 2 2 Skewness -.5474952 99% 2 2 Kurtosis 1.299751 What is your age? ------------------------------------------------------------- Percentiles Smallest 1% 18 18 5% 18 18 10% 18 18 Obs 481 25% 18 18 Sum of Wgt. 481 50% 19 Mean 19.921 Largest Std. Dev. 2.06811 75% 21 28 90% 22 29 Variance 4.277079 95% 24 30 Skewness 1.598245 99% 27 30 Kurtosis 6.552096 ALS ------------------------------------------------------------- Percentiles Smallest 1% 18 18 5% 18 18 10% 19 18 Obs 481 25% 24 18 Sum of Wgt. 481 50% 33 Mean 35.20582 Largest Std. Dev. 13.12604 75% 44 71 90% 52 72 Variance 172.293 95% 61 72 Skewness .6243659 99% 70 72 Kurtosis 2.711559 negurgtot ------------------------------------------------------------- Percentiles Smallest 1% 15 13 5% 17 15 10% 18 15 Obs 481 25% 22 15 Sum of Wgt. 481 50% 27 Mean 27.06445 Largest Std. Dev. 6.792306 75% 31 45 90% 36 45 Variance 46.13542 95% 39 45 Skewness .3794057 99% 44 45 Kurtosis 2.661998 premedtot ------------------------------------------------------------- Percentiles Smallest 1% 13 11 5% 15 11 10% 16 11 Obs 481 25% 19 13 Sum of Wgt. 481 50% 23 Mean 23.01455 Largest Std. Dev. 5.407729 75% 26 38 90% 30 39 Variance 29.24354 95% 32 40 Skewness .421346 99% 38 40 Kurtosis 3.063524 persevtot ------------------------------------------------------------- Percentiles Smallest 1% 10 10 5% 12 10 10% 13 10 Obs 481 25% 15 10 Sum of Wgt. 481 50% 19 Mean 19.50728 Largest Std. Dev. 5.787599 75% 23 37 90% 27 37 Variance 33.4963 95% 31 39 Skewness .7724914 99% 37 40 Kurtosis 3.381991 senseektot ------------------------------------------------------------- Percentiles Smallest 1% 16 12 5% 21 13 10% 24 15 Obs 481 25% 28 16 Sum of Wgt. 481 50% 33 Mean 32.97713 Largest Std. Dev. 6.881852 75% 38 47 90% 42 47 Variance 47.35989 95% 43 48 Skewness -.2714306 99% 46 48 Kurtosis 2.6015 posurgtot ------------------------------------------------------------- Percentiles Smallest 1% 20 18 5% 21 19 10% 22 19 Obs 481 25% 24 19 Sum of Wgt. 481 50% 29 Mean 30.06653 Largest Std. Dev. 7.450373 75% 34 51 90% 40 52 Variance 55.50806 95% 46 53 Skewness .8747951 99% 51 56 Kurtosis 3.335174 phys_agg ------------------------------------------------------------- Percentiles Smallest 1% 0 0 5% 0 0 10% 0 0 Obs 481 25% 0 0 Sum of Wgt. 481 50% 0 Mean 1.661123 Largest Std. Dev. 3.993683 75% 2 21 90% 5 22 Variance 15.94951 95% 10 35 Skewness 4.158179 99% 17 35 Kurtosis 26.56256 . pwcorr sex age ALS negurgtot premedtot persevtot senseektot posurgtot phys_agg, sig obs | sex age ALS negurg~t premed~t persev~t sensee~t -------------+--------------------------------------------------------------- sex | 1.0000 | | 481 | age | -0.1523 1.0000 | 0.0008 | 481 481 | ALS | 0.1208 -0.0468 1.0000 | 0.0080 0.3062 | 481 481 481 | negurgtot | 0.0416 0.0287 0.2767 1.0000 | 0.3631 0.5302 0.0000 | 481 481 481 481 | premedtot | -0.0019 0.0063 0.0294 0.3121 1.0000 | 0.9662 0.8912 0.5205 0.0000 | 481 481 481 481 481 | persevtot | -0.0367 0.0079 0.1596 0.5205 0.5540 1.0000 | 0.4220 0.8631 0.0004 0.0000 0.0000 | 481 481 481 481 481 481 | senseektot | -0.1530 0.0161 0.0215 0.2253 0.1120 -0.1281 1.0000 | 0.0008 0.7243 0.6377 0.0000 0.0140 0.0049 | 481 481 481 481 481 481 481 | posurgtot | -0.1032 -0.0352 0.1344 0.7010 0.2573 0.4677 0.2436 | 0.0236 0.4409 0.0031 0.0000 0.0000 0.0000 0.0000 | 481 481 481 481 481 481 481 | phys_agg | -0.0659 -0.0701 0.1388 0.1176 0.0998 0.0956 0.1036 | 0.1490 0.1248 0.0023 0.0098 0.0287 0.0361 0.0231 | 481 481 481 481 481 481 481 | | posurg~t phys_agg -------------+------------------ posurgtot | 1.0000 | | 481 | phys_agg | 0.1729 1.0000 | 0.0001 | 481 481 | . . * The hurdle models were estimated through a 2-step process following . * Long & Freese (2006). We first estimate a logistic model using complmentary . * log-log regression. We and then estimate a zero-truncated negative . * binomial model. The chi-square and df are simply added for the full model. . * Technically, the models can be estimated using the ml approach which results . * in an equivalent model. For an example of the ml approach see: . * (http://www.stata-journal.com/sjpdf.html?articlenum=st0040). . * However, this approach does not allow for differences in . * centering (which is needed as there are not the same number of . * observations in the count & logistic models because not . * everyone engaged in aggressive acts). . . * Note1. In the paper, we report Cragg-Uhler R2, which is based on the . * logistic analysis. You can compute a variety of different R2, but you . * cannot get a single model R2 using this approach. An alternative approach . * using the ml method followed by predict, then correlating the predicted . * value with the observed value has been suggested. We opted for a more . * conservative approach. . . * Note2. In the paper, we report Odds Ratios for the logistic . * portion of the model, and Incident Rate Ratios for the count . * portion of the model. To get these, simply exponentiate the . * coefficients in the output. . . * Step 1 . cloglog PAlog C_sex1 C_age1 , nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(2) = 12.13 Log likelihood = -292.43116 Prob > chi2 = 0.0023 ------------------------------------------------------------------------------ PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex1 | -.4615449 .1671998 -2.76 0.006 -.7892504 -.1338393 C_age1 | -.1100049 .0458834 -2.40 0.017 -.1999348 -.020075 _cons | -1.011056 .0845459 -11.96 0.000 -1.176763 -.8453494 ------------------------------------------------------------------------------ . fitstat Measures of Fit for cloglog of PAlog Log-Lik Intercept Only: -298.496 Log-Lik Full Model: -292.431 D(478): 584.862 LR(2): 12.129 Prob > LR: 0.002 McFadden's R2: 0.020 McFadden's Adj R2: 0.010 ML (Cox-Snell) R2: 0.025 Cragg-Uhler(Nagelkerke) R2: 0.035 Efron's R2: 0.024 Count R2: 0.688 Adj Count R2: 0.000 AIC: 1.228 AIC*n: 590.862 BIC: -2367.202 BIC': 0.223 BIC used by Stata: 603.390 AIC used by Stata: 590.862 . ztnb newPA sex age Fitting Zero-truncated poisson model: Iteration 0: log likelihood = -561.89618 Iteration 1: log likelihood = -561.88595 Iteration 2: log likelihood = -561.88595 Fitting constant-only model: Iteration 0: log likelihood = -388.37508 Iteration 1: log likelihood = -384.97663 Iteration 2: log likelihood = -384.67607 Iteration 3: log likelihood = -384.67406 Iteration 4: log likelihood = -384.67405 Fitting full model: Iteration 0: log likelihood = -384.59839 Iteration 1: log likelihood = -384.57883 Iteration 2: log likelihood = -384.57881 Zero-truncated negative binomial regression Number of obs = 150 LR chi2(2) = 0.19 Dispersion = mean Prob > chi2 = 0.9092 Log likelihood = -384.57881 Pseudo R2 = 0.0002 ------------------------------------------------------------------------------ newPA | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- sex | -.0236978 .2265924 -0.10 0.917 -.4678107 .4204151 age | -.0311755 .0715931 -0.44 0.663 -.1714955 .1091444 _cons | 1.987432 1.574142 1.26 0.207 -1.097829 5.072693 -------------+---------------------------------------------------------------- /lnalpha | .4186384 .3059111 -.1809363 1.018213 -------------+---------------------------------------------------------------- alpha | 1.519891 .4649514 .8344885 2.768244 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 354.61 Prob>=chibar2 = 0.000 . fitstat Measures of Fit for ztnb of newPA Log-Lik Intercept Only: -384.674 Log-Lik Full Model: -384.579 D(146): 769.158 LR(2): 0.190 Prob > LR: 0.909 McFadden's R2: 0.000 McFadden's Adj R2: -0.010 ML (Cox-Snell) R2: 0.001 Cragg-Uhler(Nagelkerke) R2: 0.001 AIC: 5.181 AIC*n: 777.158 BIC: 37.605 BIC': 9.831 BIC used by Stata: 789.200 AIC used by Stata: 777.158 . . * Step 2 . cloglog PAlog C_sex1 C_age1 C_ALS1 C_negurgtot1 C_premedtot1 C_persevtot1 C_senseektot1 C_posurgtot1 , nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(8) = 31.48 Log likelihood = -282.75801 Prob > chi2 = 0.0001 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4500158 .1747174 -2.58 0.010 -.7924556 -.1075761 C_age1 | -.1071516 .0465735 -2.30 0.021 -.198434 -.0158692 C_ALS1 | .0094583 .0061169 1.55 0.122 -.0025306 .0214473 C_negurgtot1 | .0113366 .019216 0.59 0.555 -.0263262 .0489994 C_premedtot1 | .0185303 .0193426 0.96 0.338 -.0193806 .0564411 C_persevtot1 | .0076649 .0214214 0.36 0.720 -.0343203 .0496501 C_senseektot1 | .0047092 .0134329 0.35 0.726 -.0216188 .0310372 C_posurgtot1 | .0224301 .0158111 1.42 0.156 -.008559 .0534192 _cons | -1.052705 .0881669 -11.94 0.000 -1.225509 -.8799012 ------------------------------------------------------------------------------- . fitstat Measures of Fit for cloglog of PAlog Log-Lik Intercept Only: -298.496 Log-Lik Full Model: -282.758 D(472): 565.516 LR(8): 31.475 Prob > LR: 0.000 McFadden's R2: 0.053 McFadden's Adj R2: 0.023 ML (Cox-Snell) R2: 0.063 Cragg-Uhler(Nagelkerke) R2: 0.089 Efron's R2: 0.058 Count R2: 0.678 Adj Count R2: -0.033 AIC: 1.213 AIC*n: 583.516 BIC: -2349.493 BIC': 17.932 BIC used by Stata: 621.099 AIC used by Stata: 583.516 . ztnb newPA C_sex C_age C_ALS C_negurgtot C_premedtot C_persevtot C_senseektot C_posurgtot Fitting Zero-truncated poisson model: Iteration 0: log likelihood = -517.13725 Iteration 1: log likelihood = -517.04061 Iteration 2: log likelihood = -517.04059 Fitting constant-only model: Iteration 0: log likelihood = -388.37508 Iteration 1: log likelihood = -384.97663 Iteration 2: log likelihood = -384.67607 Iteration 3: log likelihood = -384.67406 Iteration 4: log likelihood = -384.67405 Fitting full model: Iteration 0: log likelihood = -377.73243 Iteration 1: log likelihood = -376.7431 Iteration 2: log likelihood = -376.69018 Iteration 3: log likelihood = -376.69005 Iteration 4: log likelihood = -376.69005 Zero-truncated negative binomial regression Number of obs = 150 LR chi2(8) = 15.97 Dispersion = mean Prob > chi2 = 0.0428 Log likelihood = -376.69005 Pseudo R2 = 0.0208 ------------------------------------------------------------------------------ newPA | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | -.0697361 .2220361 -0.31 0.753 -.504919 .3654467 C_age | -.0421453 .0650775 -0.65 0.517 -.1696948 .0854043 C_ALS | .0266393 .008682 3.07 0.002 .0096229 .0436558 C_negurgtot | -.0400668 .0220633 -1.82 0.069 -.0833101 .0031764 C_premedtot | .0121065 .0168688 0.72 0.473 -.0209557 .0451687 C_persevtot | .0077683 .0207099 0.38 0.708 -.0328224 .0483589 C_senseektot | .0223725 .0158128 1.41 0.157 -.00862 .0533651 C_posurgtot | .0355156 .0197376 1.80 0.072 -.0031695 .0742007 _cons | 1.363637 .1215674 11.22 0.000 1.12537 1.601905 -------------+---------------------------------------------------------------- /lnalpha | .0983632 .2738518 -.4383763 .6351028 -------------+---------------------------------------------------------------- alpha | 1.103363 .302158 .645083 1.887216 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 280.70 Prob>=chibar2 = 0.000 . fitstat Measures of Fit for ztnb of newPA Log-Lik Intercept Only: -384.674 Log-Lik Full Model: -376.690 D(140): 753.380 LR(8): 15.968 Prob > LR: 0.043 McFadden's R2: 0.021 McFadden's Adj R2: -0.005 ML (Cox-Snell) R2: 0.101 Cragg-Uhler(Nagelkerke) R2: 0.102 AIC: 5.156 AIC*n: 773.380 BIC: 51.891 BIC': 24.117 BIC used by Stata: 803.486 AIC used by Stata: 773.380 . . * Step 3 . cloglog PAlog C_sex1 C_age1 C_ALS1 C_negurgtot1 C_premedtot1 C_persevtot1 C_senseektot1 C_posurgtot1 NUxALS1 PUxALS1 PRExALS1 PERxALS1 SSxALS1, nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(13) = 37.58 Log likelihood = -279.70351 Prob > chi2 = 0.0003 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4890398 .1770426 -2.76 0.006 -.8360369 -.1420427 C_age1 | -.108916 .0462913 -2.35 0.019 -.1996453 -.0181866 C_ALS1 | .0135304 .0066198 2.04 0.041 .0005558 .026505 C_negurgtot1 | -.0013913 .0196496 -0.07 0.944 -.0399038 .0371212 C_premedtot1 | .0213013 .0196879 1.08 0.279 -.0172862 .0598889 C_persevtot1 | .0076957 .0223318 0.34 0.730 -.0360738 .0514653 C_senseektot1 | .0074825 .0144298 0.52 0.604 -.0207993 .0357643 C_posurgtot1 | .0306469 .015983 1.92 0.055 -.0006792 .061973 NUxALS1 | .0030921 .001298 2.38 0.017 .000548 .0056363 PUxALS1 | -.0024303 .0012313 -1.97 0.048 -.0048437 -.0000169 PRExALS1 | -.0006784 .0014023 -0.48 0.629 -.0034268 .0020701 PERxALS1 | -.000547 .0015933 -0.34 0.731 -.0036698 .0025758 SSxALS1 | -.0003674 .0009381 -0.39 0.695 -.002206 .0014713 _cons | -1.1027 .0935535 -11.79 0.000 -1.286061 -.9193383 ------------------------------------------------------------------------------- . fitstat Measures of Fit for cloglog of PAlog Log-Lik Intercept Only: -298.496 Log-Lik Full Model: -279.704 D(467): 559.407 LR(13): 37.584 Prob > LR: 0.000 McFadden's R2: 0.063 McFadden's Adj R2: 0.016 ML (Cox-Snell) R2: 0.075 Cragg-Uhler(Nagelkerke) R2: 0.106 Efron's R2: 0.072 Count R2: 0.692 Adj Count R2: 0.013 AIC: 1.221 AIC*n: 587.407 BIC: -2324.723 BIC': 42.702 BIC used by Stata: 645.869 AIC used by Stata: 587.407 . ztnb newPA C_sex C_age C_ALS C_negurgtot C_premedtot C_persevtot C_senseektot C_posurgtot PUxALS NUxALS PRExALS PERxALS SSxALS Fitting Zero-truncated poisson model: Iteration 0: log likelihood = -509.4681 Iteration 1: log likelihood = -509.34845 Iteration 2: log likelihood = -509.34842 Fitting constant-only model: Iteration 0: log likelihood = -388.37508 Iteration 1: log likelihood = -384.97663 Iteration 2: log likelihood = -384.67607 Iteration 3: log likelihood = -384.67406 Iteration 4: log likelihood = -384.67405 Fitting full model: Iteration 0: log likelihood = -377.39117 Iteration 1: log likelihood = -376.18026 Iteration 2: log likelihood = -376.06307 Iteration 3: log likelihood = -376.06275 Iteration 4: log likelihood = -376.06275 Zero-truncated negative binomial regression Number of obs = 150 LR chi2(13) = 17.22 Dispersion = mean Prob > chi2 = 0.1893 Log likelihood = -376.06275 Pseudo R2 = 0.0224 ------------------------------------------------------------------------------ newPA | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | -.0768682 .2232854 -0.34 0.731 -.5144995 .3607631 C_age | -.0498035 .0646484 -0.77 0.441 -.176512 .076905 C_ALS | .0262354 .0087233 3.01 0.003 .009138 .0433328 C_negurgtot | -.0383396 .0227195 -1.69 0.092 -.082869 .0061897 C_premedtot | .00705 .0176253 0.40 0.689 -.027495 .0415951 C_persevtot | .0147105 .0218328 0.67 0.500 -.0280811 .057502 C_senseektot | .0183454 .0168045 1.09 0.275 -.0145909 .0512817 C_posurgtot | .0291505 .0213536 1.37 0.172 -.0127017 .0710027 PUxALS | .0008386 .001541 0.54 0.586 -.0021817 .003859 NUxALS | .000416 .0016288 0.26 0.798 -.0027764 .0036084 PRExALS | .0007458 .0014366 0.52 0.604 -.00207 .0035616 PERxALS | -.0010498 .0019351 -0.54 0.587 -.0048424 .0027429 SSxALS | .0001495 .0011875 0.13 0.900 -.0021779 .0024769 _cons | 1.363912 .1237612 11.02 0.000 1.121344 1.606479 -------------+---------------------------------------------------------------- /lnalpha | .0680881 .2723214 -.4656522 .6018283 -------------+---------------------------------------------------------------- alpha | 1.07046 .2915091 .6277256 1.825453 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 266.57 Prob>=chibar2 = 0.000 . fitstat Measures of Fit for ztnb of newPA Log-Lik Intercept Only: -384.674 Log-Lik Full Model: -376.063 D(135): 752.125 LR(13): 17.223 Prob > LR: 0.189 McFadden's R2: 0.022 McFadden's Adj R2: -0.017 ML (Cox-Snell) R2: 0.108 Cragg-Uhler(Nagelkerke) R2: 0.109 AIC: 5.214 AIC*n: 782.125 BIC: 75.690 BIC': 47.916 BIC used by Stata: 827.285 AIC used by Stata: 782.125 . . * Final Step . cloglog PAlog C_sex1 C_age1 C_ALS1 C_negurgtot1 C_premedtot1 C_persevtot1 C_senseektot1 C_posurgtot1 NUxALS1 PUxALS1, nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(10) = 36.57 Log likelihood = -280.20942 Prob > chi2 = 0.0001 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4885628 .1766135 -2.77 0.006 -.8347188 -.1424067 C_age1 | -.108323 .0463063 -2.34 0.019 -.1990816 -.0175643 C_ALS1 | .0128572 .0065685 1.96 0.050 -.0000168 .0257312 C_negurgtot1 | -.0001365 .0194635 -0.01 0.994 -.0382842 .0380113 C_premedtot1 | .0221225 .0193842 1.14 0.254 -.0158698 .0601148 C_persevtot1 | .0060447 .0215628 0.28 0.779 -.0362178 .0483071 C_senseektot1 | .0042408 .013511 0.31 0.754 -.0222403 .0307219 C_posurgtot1 | .0305977 .0159066 1.92 0.054 -.0005786 .061774 NUxALS1 | .0027852 .0012316 2.26 0.024 .0003714 .0051991 PUxALS1 | -.0025239 .0011887 -2.12 0.034 -.0048536 -.0001942 _cons | -1.0989 .0930038 -11.82 0.000 -1.281184 -.9166162 ------------------------------------------------------------------------------- . fitstat Measures of Fit for cloglog of PAlog Log-Lik Intercept Only: -298.496 Log-Lik Full Model: -280.209 D(470): 560.419 LR(10): 36.573 Prob > LR: 0.000 McFadden's R2: 0.061 McFadden's Adj R2: 0.024 ML (Cox-Snell) R2: 0.073 Cragg-Uhler(Nagelkerke) R2: 0.103 Efron's R2: 0.070 Count R2: 0.686 Adj Count R2: -0.007 AIC: 1.211 AIC*n: 582.419 BIC: -2342.239 BIC': 25.186 BIC used by Stata: 628.353 AIC used by Stata: 582.419 . ztnb newPA C_sex C_age C_ALS C_negurgtot C_premedtot C_persevtot C_senseektot C_posurgtot Fitting Zero-truncated poisson model: Iteration 0: log likelihood = -517.13725 Iteration 1: log likelihood = -517.04061 Iteration 2: log likelihood = -517.04059 Fitting constant-only model: Iteration 0: log likelihood = -388.37508 Iteration 1: log likelihood = -384.97663 Iteration 2: log likelihood = -384.67607 Iteration 3: log likelihood = -384.67406 Iteration 4: log likelihood = -384.67405 Fitting full model: Iteration 0: log likelihood = -377.73243 Iteration 1: log likelihood = -376.7431 Iteration 2: log likelihood = -376.69018 Iteration 3: log likelihood = -376.69005 Iteration 4: log likelihood = -376.69005 Zero-truncated negative binomial regression Number of obs = 150 LR chi2(8) = 15.97 Dispersion = mean Prob > chi2 = 0.0428 Log likelihood = -376.69005 Pseudo R2 = 0.0208 ------------------------------------------------------------------------------ newPA | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- C_sex | -.0697361 .2220361 -0.31 0.753 -.504919 .3654467 C_age | -.0421453 .0650775 -0.65 0.517 -.1696948 .0854043 C_ALS | .0266393 .008682 3.07 0.002 .0096229 .0436558 C_negurgtot | -.0400668 .0220633 -1.82 0.069 -.0833101 .0031764 C_premedtot | .0121065 .0168688 0.72 0.473 -.0209557 .0451687 C_persevtot | .0077683 .0207099 0.38 0.708 -.0328224 .0483589 C_senseektot | .0223725 .0158128 1.41 0.157 -.00862 .0533651 C_posurgtot | .0355156 .0197376 1.80 0.072 -.0031695 .0742007 _cons | 1.363637 .1215674 11.22 0.000 1.12537 1.601905 -------------+---------------------------------------------------------------- /lnalpha | .0983632 .2738518 -.4383763 .6351028 -------------+---------------------------------------------------------------- alpha | 1.103363 .302158 .645083 1.887216 ------------------------------------------------------------------------------ Likelihood-ratio test of alpha=0: chibar2(01) = 280.70 Prob>=chibar2 = 0.000 . fitstat Measures of Fit for ztnb of newPA Log-Lik Intercept Only: -384.674 Log-Lik Full Model: -376.690 D(140): 753.380 LR(8): 15.968 Prob > LR: 0.043 McFadden's R2: 0.021 McFadden's Adj R2: -0.005 ML (Cox-Snell) R2: 0.101 Cragg-Uhler(Nagelkerke) R2: 0.102 AIC: 5.156 AIC*n: 773.380 BIC: 51.891 BIC': 24.117 BIC used by Stata: 803.486 AIC used by Stata: 773.380 . . * Next we generate variables at +/- 1 SD on the impulsivity moderators . gen Hpu = C_posurgtot1 - 7.450373 . gen Lpu = C_posurgtot1 + 7.450373 . gen Hnu = C_negurgtot1 - 6.792306 . gen Lnu = C_negurgtot1 + 6.792306 . . * Then we generate interactions between the urgency variables & affect lability . gen HpuXals = Hpu*C_ALS1 . gen LpuXals = Lpu*C_ALS1 . gen HnuXals = Hnu*C_ALS1 . gen LnuXals = Lnu*C_ALS1 . . * Finally, we exmaine the simple slopes of affect lability at high . * and low levels of positive & negative urgency respectively . cloglog PAlog C_sex1 C_age1 C_ALS1 Hpu C_premedtot1 C_persevtot1 C_senseektot1 C_negurgtot1 NUxALS1 HpuXals, nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(10) = 36.57 Log likelihood = -280.20942 Prob > chi2 = 0.0001 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4885628 .1766135 -2.77 0.006 -.8347188 -.1424067 C_age1 | -.108323 .0463063 -2.34 0.019 -.1990816 -.0175643 C_ALS1 | -.005947 .0094706 -0.63 0.530 -.024509 .012615 Hpu | .0305977 .0159066 1.92 0.054 -.0005786 .061774 C_premedtot1 | .0221225 .0193842 1.14 0.254 -.0158698 .0601148 C_persevtot1 | .0060447 .0215628 0.28 0.779 -.0362178 .0483071 C_senseektot1 | .0042408 .013511 0.31 0.754 -.0222403 .0307219 C_negurgtot1 | -.0001365 .0194635 -0.01 0.994 -.0382842 .0380113 NUxALS1 | .0027852 .0012316 2.26 0.024 .0003714 .0051991 HpuXals | -.0025239 .0011887 -2.12 0.034 -.0048536 -.0001942 _cons | -.8709359 .138891 -6.27 0.000 -1.143157 -.5987145 ------------------------------------------------------------------------------- . cloglog PAlog C_sex1 C_age1 C_ALS1 Lpu C_premedtot1 C_persevtot1 C_senseektot1 C_negurgtot1 NUxALS1 LpuXals, nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(10) = 36.57 Log likelihood = -280.20942 Prob > chi2 = 0.0001 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4885628 .1766135 -2.77 0.006 -.8347188 -.1424067 C_age1 | -.108323 .0463063 -2.34 0.019 -.1990816 -.0175643 C_ALS1 | .0316614 .0123876 2.56 0.011 .0073821 .0559406 Lpu | .0305977 .0159066 1.92 0.054 -.0005786 .061774 C_premedtot1 | .0221225 .0193842 1.14 0.254 -.0158698 .0601148 C_persevtot1 | .0060447 .0215628 0.28 0.779 -.0362178 .0483071 C_senseektot1 | .0042408 .013511 0.31 0.754 -.0222403 .0307219 C_negurgtot1 | -.0001365 .0194635 -0.01 0.994 -.0382842 .0380113 NUxALS1 | .0027852 .0012316 2.26 0.024 .0003714 .0051991 LpuXals | -.0025239 .0011887 -2.12 0.034 -.0048536 -.0001942 _cons | -1.326864 .1615482 -8.21 0.000 -1.643493 -1.010236 ------------------------------------------------------------------------------- . cloglog PAlog C_sex1 C_age1 C_ALS1 Hnu C_premedtot1 C_persevtot1 C_senseektot1 C_posurgtot1 PUxALS1 HnuXals, nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(10) = 36.57 Log likelihood = -280.20942 Prob > chi2 = 0.0001 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4885628 .1766135 -2.77 0.006 -.8347188 -.1424067 C_age1 | -.108323 .0463063 -2.34 0.019 -.1990816 -.0175643 C_ALS1 | .0317752 .0112919 2.81 0.005 .0096436 .0539069 Hnu | -.0001365 .0194635 -0.01 0.994 -.0382842 .0380113 C_premedtot1 | .0221225 .0193842 1.14 0.254 -.0158698 .0601148 C_persevtot1 | .0060447 .0215628 0.28 0.779 -.0362178 .0483071 C_senseektot1 | .0042408 .013511 0.31 0.754 -.0222403 .0307219 C_posurgtot1 | .0305977 .0159066 1.92 0.054 -.0005786 .061774 PUxALS1 | -.0025239 .0011887 -2.12 0.034 -.0048536 -.0001942 HnuXals | .0027852 .0012316 2.26 0.024 .0003714 .0051991 _cons | -1.099827 .1654198 -6.65 0.000 -1.424044 -.7756101 ------------------------------------------------------------------------------- . cloglog PAlog C_sex1 C_age1 C_ALS1 Lnu C_premedtot1 C_persevtot1 C_senseektot1 C_posurgtot1 PUxALS1 LnuXals, nolog Complementary log-log regression Number of obs = 481 Zero outcomes = 331 Nonzero outcomes = 150 LR chi2(10) = 36.57 Log likelihood = -280.20942 Prob > chi2 = 0.0001 ------------------------------------------------------------------------------- PAlog | Coef. Std. Err. z P>|z| [95% Conf. Interval] --------------+---------------------------------------------------------------- C_sex1 | -.4885628 .1766135 -2.77 0.006 -.8347188 -.1424067 C_age1 | -.108323 .0463063 -2.34 0.019 -.1990816 -.0175643 C_ALS1 | -.0060609 .0099368 -0.61 0.542 -.0255367 .013415 Lnu | -.0001365 .0194635 -0.01 0.994 -.0382842 .0380113 C_premedtot1 | .0221225 .0193842 1.14 0.254 -.0158698 .0601148 C_persevtot1 | .0060447 .0215628 0.28 0.779 -.0362178 .0483071 C_senseektot1 | .0042408 .013511 0.31 0.754 -.0222403 .0307219 C_posurgtot1 | .0305977 .0159066 1.92 0.054 -.0005786 .061774 PUxALS1 | -.0025239 .0011887 -2.12 0.034 -.0048536 -.0001942 LnuXals | .0027852 .0012316 2.26 0.024 .0003714 .0051991 _cons | -1.097973 .1577669 -6.96 0.000 -1.407191 -.7887558 ------------------------------------------------------------------------------- . . . . end of do-file . log close name: log: C:\Users\Owner\Desktop\AggIll\ab_upps_als.smcl log type: smcl closed on: 23 Nov 2012, 12:05:36 ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------