------------------------------------------------------------------------------- name: log: Q:\C-modelling\runmlwin\website\logfiles\2020-03-27\16\12.5.smcl log type: smcl opened on: 27 Mar 2020, 18:24:05 . **************************************************************************** . * Module 12: Cross-Classified Models - Stata Practical . * . * P12.5: Adding Predictor Variables . * . * George Leckie . * Centre for Multilevel Modelling, 2011 . **************************************************************************** . * Stata do-file to replicate all analyses using runmlwin . * . * George Leckie . * Centre for Multilevel Modelling, 2013 . * http://www.bristol.ac.uk/cmm/software/runmlwin/ . **************************************************************************** . . * P12.5.1 Adding student level predictor variables . . use "http://www.bristol.ac.uk/cmm/media/runmlwin/12.5.dta", clear . . xtmixed attain p7vrq p7read || _all: R.schid || neighid:, mle variance Performing EM optimization: Performing gradient-based optimization: Iteration 0: log likelihood = -2460.2597 Iteration 1: log likelihood = -2459.1782 Iteration 2: log likelihood = -2459.1764 Iteration 3: log likelihood = -2459.1764 Computing standard errors: Mixed-effects ML regression Number of obs = 2,310 ------------------------------------------------------------- | No. of Observations per Group Group Variable | Groups Minimum Average Maximum ----------------+-------------------------------------------- _all | 1 2,310 2,310.0 2,310 neighid | 524 1 4.4 16 ------------------------------------------------------------- Wald chi2(2) = 2104.33 Log likelihood = -2459.1764 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ attain | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- p7vrq | .0302058 .0023145 13.05 0.000 .0256694 .0347422 p7read | .0293516 .0017798 16.49 0.000 .0258633 .0328398 _cons | .0818216 .0247908 3.30 0.001 .0332324 .1304108 ------------------------------------------------------------------------------ ------------------------------------------------------------------------------ Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval] -----------------------------+------------------------------------------------ _all: Identity | var(R.schid) | .0054713 .0036475 .0014812 .0202096 -----------------------------+------------------------------------------------ neighid: Identity | var(_cons) | .0269212 .008632 .0143604 .0504688 -----------------------------+------------------------------------------------ var(Residual) | .4656259 .0152837 .4366136 .496566 ------------------------------------------------------------------------------ LR test vs. linear model: chi2(2) = 21.33 Prob > chi2 = 0.0000 Note: LR test is conservative and provided only for reference. . . estimates store model4 . . lrtest model1 model4 Likelihood-ratio test LR chi2(2) = 1438.36 (Assumption: model1 nested in model4) Prob > chi2 = 0.0000 . . test p7vrq p7read ( 1) [attain]p7vrq = 0 ( 2) [attain]p7read = 0 chi2( 2) = 2104.33 Prob > chi2 = 0.0000 . . xtmixed attain p7vrq p7read male dadocc daded momed dadunemp /// > || _all: R.schid || neighid:, mle variance Performing EM optimization: Performing gradient-based optimization: Iteration 0: log likelihood = -2404.9248 Iteration 1: log likelihood = -2402.297 Iteration 2: log likelihood = -2402.2937 Iteration 3: log likelihood = -2402.2937 Computing standard errors: Mixed-effects ML regression Number of obs = 2,310 ------------------------------------------------------------- | No. of Observations per Group Group Variable | Groups Minimum Average Maximum ----------------+-------------------------------------------- _all | 1 2,310 2,310.0 2,310 neighid | 524 1 4.4 16 ------------------------------------------------------------- Wald chi2(7) = 2415.24 Log likelihood = -2402.2937 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ attain | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- p7vrq | .02823 .0022758 12.40 0.000 .0237696 .0326904 p7read | .0269051 .0017587 15.30 0.000 .0234581 .0303521 male | -.0540241 .0285779 -1.89 0.059 -.1100358 .0019875 dadocc | .0091773 .0013584 6.76 0.000 .0065148 .0118398 daded | .1487033 .0410774 3.62 0.000 .068193 .2292136 momed | .0649316 .0376491 1.72 0.085 -.0088593 .1387225 dadunemp | -.1464694 .0469 -3.12 0.002 -.2383916 -.0545471 _cons | .0805349 .0272663 2.95 0.003 .027094 .1339758 ------------------------------------------------------------------------------ ------------------------------------------------------------------------------ Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval] -----------------------------+------------------------------------------------ _all: Identity | var(R.schid) | .0030572 .0024765 .0006249 .0149566 -----------------------------+------------------------------------------------ neighid: Identity | var(_cons) | .0122113 .0072414 .0038194 .0390422 -----------------------------+------------------------------------------------ var(Residual) | .4551568 .0148446 .4269722 .4852018 ------------------------------------------------------------------------------ LR test vs. linear model: chi2(2) = 7.40 Prob > chi2 = 0.0248 Note: LR test is conservative and provided only for reference. . . estimates store model5 . . lrtest model4 model5 Likelihood-ratio test LR chi2(5) = 113.77 (Assumption: model4 nested in model5) Prob > chi2 = 0.0000 . . . . * P12.5.2 Adding neighbourhood level predictor variables . . xtmixed attain p7vrq p7read male dadocc daded momed dadunemp /// > deprive /// > || _all: R.schid || neighid:, mle variance Performing EM optimization: Performing gradient-based optimization: Iteration 0: log likelihood = -2389.0906 Iteration 1: log likelihood = -2384.6727 Iteration 2: log likelihood = -2384.6678 Iteration 3: log likelihood = -2384.6678 Computing standard errors: Mixed-effects ML regression Number of obs = 2,310 ------------------------------------------------------------- | No. of Observations per Group Group Variable | Groups Minimum Average Maximum ----------------+-------------------------------------------- _all | 1 2,310 2,310.0 2,310 neighid | 524 1 4.4 16 ------------------------------------------------------------- Wald chi2(8) = 2525.72 Log likelihood = -2384.6678 Prob > chi2 = 0.0000 ------------------------------------------------------------------------------ attain | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- p7vrq | .0275636 .002263 12.18 0.000 .0231282 .031999 p7read | .0262471 .00175 15.00 0.000 .0228172 .029677 male | -.0559606 .0283915 -1.97 0.049 -.1116069 -.0003142 dadocc | .0081125 .0013604 5.96 0.000 .0054462 .0107789 daded | .143641 .0407871 3.52 0.000 .0636998 .2235821 momed | .0594877 .0373803 1.59 0.112 -.0137763 .1327517 dadunemp | -.1207028 .0467775 -2.58 0.010 -.212385 -.0290206 deprive | -.1561175 .0255825 -6.10 0.000 -.2062582 -.1059768 _cons | .0856904 .0276423 3.10 0.002 .0315125 .1398684 ------------------------------------------------------------------------------ ------------------------------------------------------------------------------ Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval] -----------------------------+------------------------------------------------ _all: Identity | var(R.schid) | .0038022 .0025793 .001006 .0143699 -----------------------------+------------------------------------------------ neighid: Identity | var(_cons) | .0035229 .0066884 .0000853 .1455218 -----------------------------+------------------------------------------------ var(Residual) | .455632 .0148495 .4274376 .4856862 ------------------------------------------------------------------------------ LR test vs. linear model: chi2(2) = 6.57 Prob > chi2 = 0.0374 Note: LR test is conservative and provided only for reference. . . estimates store model6 . . lrtest model5 model6 Likelihood-ratio test LR chi2(1) = 35.25 (Assumption: model5 nested in model6) Prob > chi2 = 0.0000 . . . . * P12.5.3 Adding school level predictor variables . end of do-file