UCLA: Statistical Consulting Groupã®ãã¼ã¸ããã
UCLAã§ã¯Mplusã®ã³ã¼ããæ¸ããã¦ãããããã®ã¨ã³ããªã§ã¯ãåãåæãRã®lavaanã§ã®åç¾ãããã¨æãã
Mplus
ãã¹è§£æã¯ãã¹ã¦ã®å¤æ°ã観測ãããæ¹ç¨å¼ç³»ãæ¨å®ããããã«ä½¿ç¨ããããæ½å¨å¤æ°ãå«ãã¢ãã«ã¨ã¯ç°ãªãããã¹ã»ã¢ãã«ã¯ï¼è¦³å¯ãããå¤æ°ã®å®å ¨ãªæ¸¬å®ãä»®å®ããã観å¯ãããå¤æ°éã®æ§é çé¢ä¿ã®ã¿ãã¢ãã«åãããããã®ã¿ã¤ãã®ã¢ãã«ã¯ã1ã¤ã¾ãã¯è¤æ°ã®å¤æ°ãä»ã®2ã¤ã®å¤æ°ã®é¢ä¿ãåªä»ããã¨èããããå ´åã«ãã使ç¨ããã¾ãï¼åªä»ã¢ãã«ï¼ãåæ§ã®ã¢ãã«è¨å®ã¯ã2ã¤ã®ç¡é¢ä¿ãªå¾å±å¤æ°ã®èª¤å·®ï¼æ®å·®ï¼ãç¸é¢ãããã¨ã許ãããã¢ãã«ï¼ä¸è¦ç¡é¢é£å帰ï¼ããå¤æ°éã®é¢ä¿ãã°ã«ã¼ãéã§ç°ãªãã¨èããããã¢ãã«ï¼å¤éã°ã«ã¼ãã¢ãã«ï¼ã®æ¨å®ã«ä½¿ç¨ãããã¨ãã§ããã
1.ç¹å®ã¢ãã«
ãã®ãã¼ã¸ã®ä¾ã§ã¯ãåçè ã®é«æ ¡æ代ã®æ績 (hs), 大å¦æ代ã®æ績 (col), GRE ã®ã¹ã³ã¢ (gre), ããã¦å¤§å¦é¢æ代ã®æ績 (grad) ã¨ãã 4 ã¤ã®å¤æ°ãå«ããã¼ã¿ã»ãã (https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat) ã使ç¨ãã¦ãããGREã¹ã³ã¢ã¯é«æ ¡ã¨å¤§å¦ã®GPAï¼ããããhsã¨colï¼ã使ã£ã¦äºæ¸¬ãã大å¦é¢ã®GPAï¼gradï¼ã¯GREãé«æ ¡GPAã大å¦GPAã使ã£ã¦äºæ¸¬ãã¾ãããã®ã¢ãã«ã¯ãã¡ããã©åå®ãããèªç±åº¦ãã¼ãã§ãããã¨ãæå³ããã
model: ã³ãã³ãã§ã¯ãon ã¨ãããã¼ã¯ã¼ã㧠gre ã hs 㨠col ã«ãgrad ã hsãcolãgre ã«å帰ããããã¨ã表ãã¦ãã¾ããoutput: ã³ãã³ãã« stdyx; ãªãã·ã§ã³ãã¤ããã¨ãæ¨æºåãããå帰ä¿æ°ã¨R2ä¹ã®å¤ãå¾ãããï¼stdyx; ãªãã·ã§ã³ã¯ãå帰ä¿æ°ã¨R2ä¹ã®å¤ãå¾ããã¾ãï¼ã(stdyx;ãªãã·ã§ã³ã¯yã¨xã®ä¸¡æ¹ã§æ¨æºåãããä¿æ°ãçæããããä»ã®ã¿ã¤ãã®æ¨æºåãå¯è½ã§ãstandardized;ãªãã·ã§ã³ã使ã£ã¦è¦æ±ã§ãã)ã
Title: Path analysis -- just identified model Data: file is https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat ; Variable: Names are hs gre col grad; Model: gre on hs col; grad on hs col gre; Output: stdyx;
以ä¸ã¯ãMplusããã®åºåã§ããã
INPUT READING TERMINATED NORMALLY Path analysis -- just identified model SUMMARY OF ANALYSIS Number of groups 1 Number of observations 200 Number of dependent variables 2 Number of independent variables 2 Number of continuous latent variables 0 Observed dependent variables Continuous GRE GRAD Observed independent variables HS COL Estimator ML Information matrix OBSERVED Maximum number of iterations 1000 Convergence criterion 0.500D-04 Maximum number of steepest descent iterations 20 Input data file(s) https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat Input data format FREE THE MODEL ESTIMATION TERMINATED NORMALLY TESTS OF MODEL FIT Chi-Square Test of Model Fit Value 0.000 Degrees of Freedom 0 P-Value 0.0000 Chi-Square Test of Model Fit for the Baseline Model Value 247.004 Degrees of Freedom 5 P-Value 0.0000 CFI/TLI CFI 1.000 TLI 1.000 Loglikelihood H0 Value -2789.415 H1 Value -2789.415 Information Criteria Number of Free Parameters 9 Akaike (AIC) 5596.830 Bayesian (BIC) 5626.515 Sample-Size Adjusted BIC 5598.002 (n* = (n + 2) / 24) RMSEA (Root Mean Square Error Of Approximation) Estimate 0.000 90 Percent C.I. 0.000 0.000 Probability RMSEA <= .05 0.000 SRMR (Standardized Root Mean Square Residual) Value 0.000 MODEL RESULTS Two-Tailed Estimate S.E. Est./S.E. P-Value GRE ON HS 0.309 0.065 4.756 0.000 COL 0.400 0.071 5.625 0.000 GRAD ON HS 0.372 0.075 4.937 0.000 COL 0.123 0.084 1.465 0.143 GRE 0.369 0.078 4.754 0.000 Intercepts GRE 15.534 2.995 5.186 0.000 GRAD 6.971 3.506 1.989 0.047 Residual Variances GRE 49.694 4.969 10.000 0.000 GRAD 59.998 6.000 10.000 0.000 STANDARDIZED MODEL RESULTS STDYX Standardization Two-Tailed Estimate S.E. Est./S.E. P-Value GRE ON HS 0.335 0.068 4.887 0.000 COL 0.396 0.068 5.859 0.000 GRAD ON HS 0.356 0.070 5.073 0.000 COL 0.108 0.073 1.467 0.142 GRE 0.326 0.067 4.869 0.000 Intercepts GRE 1.643 0.378 4.343 0.000 GRAD 0.651 0.350 1.859 0.063 Residual Variances GRE 0.556 0.052 10.611 0.000 GRAD 0.523 0.051 10.240 0.000 R-SQUARE Observed Two-Tailed Variable Estimate S.E. Est./S.E. P-Value GRE 0.444 0.052 8.477 0.000 GRAD 0.477 0.051 9.333 0.000 QUALITY OF NUMERICAL RESULTS Condition Number for the Information Matrix 0.348E-04 (ratio of smallest to largest eigenvalue)
MODEL RESULTS ã®ä¸ã«ã¯ãgre ã® hs 㨠col ã¸ã®å帰ã®ãã¹ä¿æ°ï¼ã¹ãã¼ãï¼ããã㦠grad ã® hs ã¸ã®å帰ã®ãã¹ä¿æ°ã表示ããã¦ãã¾ããæ¨æºåããã¦ããªãä¿æ°ï¼Estimateã¨æ¸ãããåï¼ã¨å ±ã«ãæ¨æºèª¤å·®ï¼S.Eï¼ãä¿æ°ãæ¨æºèª¤å·®ã§å²ã£ãå¤ãããã¦på¤ã示ããã¦ããããããããhsã¨colã¯greãææã«äºæ¸¬ããgreã¨hsï¼colã¯äºæ¸¬ããï¼ã¯gradãææã«äºæ¸¬ãããã¨ãããããã¢ãã«ããã®è¿½å ãã©ã¡ã¼ã¿ã¯ããã¹ä¿æ°ã®ä¸ã«è¨è¼ããã¦ãããããã¯ããã¹ã¦ã®ä¿æ°ï¼åçã¨å¾ãï¼ãä¸ç·ã«è¡¨ç¤ºãããããã¤ãã®æ±ç¨çµ±è¨ããã±ã¼ã¸ã¨ã¯ç°ãªããoutput: ã³ãã³ãã®stdyxãªãã·ã§ã³ã使ã£ã¦æ¨æºåä¿æ°ãè¦æ±ããã®ã§ãæ¨æºåçµæãï¼éæ¨æºåçµæã®å¾ã«ï¼åºåã«å«ã¾ãã¾ããSTDYXæ¨æºåã¨ããè¦åºãã®ä¸ã«ã1åä½ã®å¤åãå ã®å¤æ°ã®æ¨æºåå·®ã®å¤åã表ãããã«ï¼æ¨æºåå帰ã¢ãã«ã¨åãããã«ï¼æ¨æºåãããã¢ãã«ã»ãã©ã¡ã¼ã¿ããã¹ã¦ãªã¹ãã¢ããããã¦ãããæ¨æºååºåã®ä¸é¨ã¨ãã¦ãR2ä¹ã®å¤ãR-SQUAREã®è¦åºãã®ä¸ã«è¡¨ç¤ºããããããã§ã¯ãæã ã®ã¢ãã«ã®åå¾å±å¤æ°ã®æ¨å®R2ä¹å¤ããæ¨æºèª¤å·®ã¨ä»®èª¬æ¤å®ã¨ã¨ãã«ä¸ãããã¦ããã
2.éæ¥å¹æããã³å ¨ä½å¹æ
ãã¹ã¢ãã«ã®é åã®1ã¤ã¯ï¼éæ¥å¹æã ãã§ãªãï¼å ¨ä½å¹æï¼ããªãã¡ï¼å¤æ°éã®é¢ä¿ï¼ãè©ä¾¡ãããã¨ãã§ããç¹ã§ããï¼å ¨ä½å¹æã¨ã¯ãç´æ¥å¹æã¨éæ¥å¹æã®çµã¿åããã§ãããã¨ã«æ³¨æããã®ä¾ã§ã¯ãhsã®gradã¸ã®éæ¥å¹æã®æ¨å®ãæ±ãã¾ãï¼greãéãã¦ï¼ãä¸å³ã¯ããã®ã¢ãã«ã«å¯¾å¿ããå³ã§ãããå¸æããéæ¥å¹æã¯éè²ã§ç¤ºããã¦ãããå ¥åãã¡ã¤ã«ã« model indirect: ã³ãã³ãã追å ããgrad ind hs; ãæå®ãããã¨ã§ãéæ¥å¹æã®æ¨å®å¤ãå¾ããã¨ãã§ããã
Title: Path analysis -- with indirect effects. Data: file is https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat ; Variable: Names are hs gre col grad; Model: gre on hs col; grad on hs col gre; Model indirect: grad ind hs; Output: stdyx;
ãã®ã¢ãã«ã®åºåã¯ä»¥ä¸ã®ã¨ããã§ããããã®ã¢ãã«ã®åºåã¯ãå ¨ä½å¹æãéæ¥å¹æãç´æ¥å¹æã示ãã»ã¯ã·ã§ã³ã追å ããã¦ãã以å¤ã¯ãåã®ã¢ãã«ã¨åããªã®ã§ãåºåã®ä¸é¨ãçç¥ãã¦ãããéæ¥å¹æã®è¿½å ã«ãããMplusãã追å ã®åºåãè¦æ±ãããããã¢ãã«èªä½ã¯å¤ãããªãã®ã§ãåºåã¯åãã§ããåè¨ãéæ¥ãããã³ç´æ¥å¹æã®å 訳ã¯ãåè¨ãåè¨éæ¥ãç¹æ®éæ¥ãããã³ç´æ¥å¹æã¨ã©ãã«ä»ããããã»ã¯ã·ã§ã³ã§ãã¢ãã«çµæã¨æ¨æºåã¢ãã«çµæã®ä¸ã«è¡¨ç¤ºããããæ¨æºåãããä¿æ°ãè¦æ±ããããããæ¨æºåãããç·å¹æãéæ¥å¹æãç´æ¥å¹æãæ¨æºåããã¦ããªãå¹æã®ä¸ã«è¡¨ç¤ºããã¦ããã
MODEL RESULTS Two-Tailed Estimate S.E. Est./S.E. P-Value GRE ON HS 0.309 0.065 4.756 0.000 COL 0.400 0.071 5.625 0.000 GRAD ON HS 0.372 0.075 4.937 0.000 COL 0.123 0.084 1.465 0.143 GRE 0.369 0.078 4.754 0.000 Intercepts GRE 15.534 2.995 5.186 0.000 GRAD 6.971 3.506 1.989 0.047 Residual Variances GRE 49.694 4.969 10.000 0.000 GRAD 59.998 6.000 10.000 0.000 <output omitted> QUALITY OF NUMERICAL RESULTS Condition Number for the Information Matrix 0.348E-04 (ratio of smallest to largest eigenvalue) TOTAL, TOTAL INDIRECT, SPECIFIC INDIRECT, AND DIRECT EFFECTS Two-Tailed Estimate S.E. Est./S.E. P-Value Effects from HS to GRAD Total 0.487 0.075 6.453 0.000 Total indirect 0.114 0.034 3.362 0.001 Specific indirect GRAD GRE HS 0.114 0.034 3.362 0.001 Direct GRAD HS 0.372 0.075 4.937 0.000 STANDARDIZED TOTAL, TOTAL INDIRECT, SPECIFIC INDIRECT, AND DIRECT EFFECTS STDYX Standardization Two-Tailed Estimate S.E. Est./S.E. P-Value Effects from HS to GRAD Total 0.465 0.068 6.858 0.000 Total indirect 0.109 0.032 3.455 0.001 Specific indirect GRAD GRE HS 0.109 0.032 3.455 0.001 Direct GRAD HS 0.356 0.070 5.073 0.000
å ·ä½çãªéæ¥å¹æã§ã¯ãGRAD GRE HS ã¨æ¸ãããå¹æï¼ãããããç¬ç«ããè¡ã«è¡¨ç¤ºãããæçµçµæãæåã«è¡¨ç¤ºããããã¨ã«æ³¨æï¼ã¯ãGREï¼ä¸ã®éããã¹ï¼ãéãã¦ãæ°åã«å¯¾ãã HS ã®éæ¥å¹æã«å¯¾ããæ¨å®ä¿æ°ã示ãã¦ãããç´æ¥å¹æã¨æ¸ãããä¿æ°ã¯ãhsãå¦ä½ã«ä¸ããç´æ¥çãªå¹æã§ãããããããhsããgradã¸ã®ææãªç´æ¥çµè·¯ã¯ãé¨åçãªä»²ä»ã«éããªããã¨ã示åãã¦ããã
3.å ·ä½çãªéæ¥å¹æ
ä¸è¨ã®ä¾ã§ã¯ãéæ¥å¹æã1ã¤ãããªããããåç´ã«éãããå¤ãã®å ´åãã¢ãã«ã«ã¯è¤æ°ã®éæ¥å¹æãããããã®ä¾ã§ã¯ãhsããcolã¸ã®æ¹åæ§ãã¹ï¼ã¤ã¾ãå帰ï¼ãé ç½®ããè¤æ°ã®éæ¥å¹æã®å¯è½æ§ãããã¢ãã«ãä½æãããä¸ã®å³ã¯ãã¢ãã«ã示ãã¦ãããæ¤è¨ããã3ã¤ã®éæ¥ãã¹ãè²ä»ãã®ç·ã§å¼·èª¿ãã¦ããã
éæ¥å¹æã®è¨ç®ãè¦æ±ããæ¹æ³ã¯ããã¤ããããæåã®æ¹æ³ã¯ãåã®ä¾ã§ç¤ºãããã®ï¼ããªãã¡ãgrad ind hs;ï¼ã§ãhs ãã grad ã¾ã§ã®ãã¹ã¦ã®éæ¥ãã¹ãè¦æ±ãããã®ã§ãããã¾ããindã使ã£ã¦ç¹å®ã®éæ¥ãã¹ãè¦æ±ãããã¨ãã§ãã¾ããä¾ãã°ã以ä¸ã§ã¯grad ind col hs;ã使ã£ã¦ãhsâcolâgradã®éæ¥å¹æï¼ã¤ã¾ããä¸å³ã®ãªã¬ã³ã¸ã®ç ´ç·ã®ãã¹ï¼ãæ¨å®ããããã¨ãæå®ãã¦ãããæå¾ã«ãvia ã使ã£ã¦ã第3ã®å¤æ°ãéããã¹ã¦ã®éæ¥å¹æãè¦æ±ã§ãããä¾ãã°ã以ä¸ã§ã¯ãgrad via gre hs; ã使ã£ã¦ãhs ãã grad ã¸ã®ãã¹ã¦ã®éæ¥ãã¹ã§ gre ãå«ããã®ãè¦æ±ãããããã¯ãhs ãã gre ãã gradï¼ã¤ã¾ããéã®å®ç·ã®ãã¹ï¼ãhs ãã col ãã gre ãã gradï¼ã¤ã¾ãããã³ã¯ã®ç¹ç·ã®ãã¹ï¼ã§ãããæ°ããæ¹åãã¹ (col on hs;) ã¨ãã¢ãã«éæ¥ã®ç¹å®ã®éæ¥ (grad ind col hs;) ã¨çµç± (grad via gre hs;) ãªãã·ã§ã³ã¯ãä¸å³ã®å ¥åã§å¼·èª¿è¡¨ç¤ºããã¦ããã
Title: Multiple indirect paths Data: file is https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat ; Variable: Names are hs gre col grad; Model: gre on col hs; grad on hs col gre; col on hs; Model indirect: grad ind col hs; grad via gre hs;
ç°¡ç¥åããåºåã¯ä»¥ä¸ããã®ã¢ãã«ã®åºåã¯ãéæ¥å¹æã示ãã»ã¯ã·ã§ã³ã追å ããã¦ãã以å¤ã¯ã以åã®ã¢ãã«ã®åºåã¨åæ§ã®æ§é ã§ãããã¨ã«æ³¨æã
<output omitted> TOTAL, TOTAL INDIRECT, SPECIFIC INDIRECT, AND DIRECT EFFECTS Two-Tailed Estimate S.E. Est./S.E. P-Value Effects from HS to GRAD Sum of indirect 0.075 0.051 1.455 0.146 Specific indirect GRAD COL HS 0.075 0.051 1.455 0.146 Effects from HS to GRAD via GRE Sum of indirect 0.204 0.047 4.333 0.000 Specific indirect GRAD GRE HS 0.114 0.034 3.362 0.001 GRAD GRE COL HS 0.090 0.026 3.487 0.000
éæ¥å¹æã®æåã®ã»ããï¼HSããGRADã¸ã®å¹æï¼ã§ã¯ãhs ã® col ãéãã grad ã«å¯¾ããéæ¥å¹æã示ããã¦ããããã®ã¢ãã«ã§ã¯ãhsã®gradã¸ã®ç´æ¥å¹æãæ¨å®ããããç¹å®ã®éæ¥å¹æãæ±ããããããã®é¨åã«ã¯è¡¨ç¤ºããã¦ããªãï¼ä¸ã«è¡¨ç¤ºããã¦ããï¼ãéæ¥å¹æã®2çªç®ã®ã»ããï¼HSããGREãçµç±ããGRADã¸ã®å¹æï¼ã¯ãGREãå«ãhsããgradã¸ã®ãã¹ã¦ã®éæ¥å¹æï¼ãã®å ´åã2ã¤ã®éæ¥å¹æãããï¼ã示ãã¦ããããã®åºåã¯ãhsãgradã«å¯¾ãã¦å ¨ä½ã¨ãã¦ææãªéæ¥å¹æï¼éæ¥å¹æã®åï¼ãæã¡ãããã«2ã¤ã®ç¹å®ã®éæ¥å¹æï¼greçµç±ãcolã¨greçµç±ï¼ãæã£ã¦ãããã¨ã示ãã¦ããããã®åºåã¯ãhsã«å¯¾ããgradã®å ¨å¹æãå«ãã§ããªããã¨ã«æ³¨æããã®åºåã§ã¯ãåã®ã¢ãã«ã§è¡ã£ãããã«ãgrad ind hsã¨æå®ããã ãã§ããã
éå°åå®ã¢ãã«
ããã¯éå°åå®ã¢ãã«ã®ä¾ã§ãæ£ã®èªç±åº¦ãæã¤ã¢ãã«ã§ããï¼é£½åã¾ãã¯åã«åå®ãããã¨è¡¨ç¾ã§ãã以åã®ã¢ãã«ã¨ã¯å¯¾ç §çã§ããï¼ãæ£ã®èªç±åº¦ãæã¤ãã¨ã§ãCFIãRMSEAãªã©ã®é©åææ¨ã¨ã¨ãã«ãã¢ãã«é©åã®ã«ã¤2ä¹æ¤å®ã使ç¨ãã¦ãã¢ãã«ã®é©åæ§ã調æ»ãããã¨ãã§ãããä¸ã®å³ã§ã¯ãã¢ãã«ã«å«ã¾ãããã¹ã¯å®ç·ã§è¡¨ãããæ¨å®ã§ãããããã§ãªããã¹ã¯ç¹ç·ã§è¡¨ããã¦ãããhsã¯gradã«ãgreã«ãç´æ¥çãªå½±é¿ãä¸ãããcolãçµç±ãã¦ã®ã¿å½±é¿ãä¸ãããã¨ã«æ³¨æãããããããã¯ãé«æ ¡ã® GPA ã¯å¤§å¦ã® GPA ã¨ã®é¢ä¿ãéãã¦ã®ã¿ãGRE ã®å¾ç¹ã大å¦é¢ã®æ績ã¨é¢é£ãããã¨ãã仮説ã«å¯¾å¿ãããã®ã§ããã
Title: Path analysis -- over identified model Data: file is https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat ; Variable: Names are hs gre col grad; Model: col on hs; gre on col; grad on col gre; Output: stdyx;
çµæã
INPUT READING TERMINATED NORMALLY Path analysis -- over identified model SUMMARY OF ANALYSIS Number of groups 1 Number of observations 200 Number of dependent variables 3 Number of independent variables 1 Number of continuous latent variables 0 Observed dependent variables Continuous GRE COL GRAD Observed independent variables HS Estimator ML Information matrix OBSERVED Maximum number of iterations 1000 Convergence criterion 0.500D-04 Maximum number of steepest descent iterations 20 Input data file(s) https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat Input data format FREE THE MODEL ESTIMATION TERMINATED NORMALLY TESTS OF MODEL FIT Chi-Square Test of Model Fit Value 44.429 Degrees of Freedom 2 P-Value 0.0000 Chi-Square Test of Model Fit for the Baseline Model Value 362.474 Degrees of Freedom 6 P-Value 0.0000 CFI/TLI CFI 0.881 TLI 0.643 Loglikelihood H0 Value -2811.629 H1 Value -2789.415 Information Criteria Number of Free Parameters 10 Akaike (AIC) 5643.258 Bayesian (BIC) 5676.242 Sample-Size Adjusted BIC 5644.561 (n* = (n + 2) / 24) RMSEA (Root Mean Square Error Of Approximation) Estimate 0.3266 90 Percent C.I. 0.247 0.412 Probability RMSEA <= .05 0.000 SRMR (Standardized Root Mean Square Residual) Value 0.086 MODEL RESULTS Two-Tailed Estimate S.E. Est./S.E. P-Value COL ON HS 0.605 0.048 12.500 0.000 GRE ON COL 0.625 0.056 11.101 0.000 GRAD ON COL 0.317 0.079 4.014 0.000 GRE 0.492 0.078 6.303 0.000 Intercepts GRE 19.887 3.009 6.609 0.000 COL 21.038 2.576 8.165 0.000 GRAD 9.779 3.664 2.669 0.008 Residual Variances GRE 55.313 5.531 10.000 0.000 COL 49.025 4.903 10.000 0.000 GRAD 67.311 6.731 10.000 0.000 STANDARDIZED MODEL RESULTS STDYX Standardization Two-Tailed Estimate S.E. Est./S.E. P-Value COL ON HS 0.662 0.040 16.684 0.000 GRE ON COL 0.617 0.044 14.112 0.000 GRAD ON COL 0.276 0.068 4.092 0.000 GRE 0.434 0.065 6.671 0.000 Intercepts GRE 2.103 0.397 5.298 0.000 COL 2.251 0.363 6.210 0.000 GRAD 0.913 0.375 2.436 0.015 Residual Variances GRE 0.619 0.054 11.452 0.000 COL 0.561 0.053 10.677 0.000 GRAD 0.587 0.053 11.002 0.000 R-SQUARE Observed Two-Tailed Variable Estimate S.E. Est./S.E. P-Value GRE 0.381 0.054 7.056 0.000 COL 0.439 0.053 8.342 0.000 GRAD 0.413 0.053 7.743 0.000 QUALITY OF NUMERICAL RESULTS Condition Number for the Information Matrix 0.104E-03 (ratio of smallest to largest eigenvalue)
ã«ã¤äºä¹ã®å¤ã¯ãç¾å¨ã®ã¢ãã«ã¨é£½åããã¢ãã«ãæ¯è¼ãããæã ã®ã¢ãã«ã¯é£½åãã¦ããªãï¼ããªãã¡ãæã ã®ã¢ãã«ã¯æ£ã®èªç±åº¦ãæã¤ï¼ã®ã§ãã«ã¤2ä¹å¤ã¯ãã¯ãã¼ãã§ã¯ãªããã¢ãã«ã®é©åæ§ãè©ä¾¡ããããã«ä½¿ç¨ãããã¨ãã§ãããåæ§ã«ãåå®ãããã°ããã®ã¢ãã«ã§1ã«çããã£ãCFIã¨TLIã¯ãç¾å¨ãæ å ±éãæã¤å¤ã«ãªã£ã¦ãããããã«ãRMSEAã¨SRMRã¯ãæ å ±éã®å¤ãå¤ãã¨ãããã«ãªãã¾ããï¼åå®ãããã°ããã®ã¢ãã«ã§ã¯ããããã¯ã¼ãã¨ãã¦è¡¨ç¤ºãããï¼ãæ£ã®èªç±åº¦ãæã¡ããããã£ã¦é©å度ææ¨ã«æçãªå¤ãæã¤ãã¨ã§ãæã ã®ã¢ãã«ãã©ã®ç¨åº¦ãã¼ã¿ã«é©åãã¦ããããããããè©ä¾¡ãããã¨ãã§ããããã®ã¢ãã«ããã®å ·ä½çãªä¿æ°æ¨å®å¤ã¯ãä¸è¬ã«ãåå®ãããã°ããã®ã¢ãã«ã¨åãããã«è§£éãããã
R
ãã¼ã¿èªã¿è¾¼ã¿
df1 <- read.table('https://stats.idre.ucla.edu/wp-content/uploads/2016/02/path.dat', sep=",") colnames(df1)<-c("hs","gre","col","grad")
1.ç¹å®ã¢ãã«
library(lavaan) model1 <- ' gre ~ hs gre ~ col grad ~ hs grad ~ col grad ~ gre ' fit1 <- sem(model = model1, data=df1) summary(fit1, standardized=TRUE)
çµæã
Regressions:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
gre ~
hs 0.309 0.065 4.756 0.000 0.309 0.335
col 0.400 0.071 5.626 0.000 0.400 0.396
grad ~
hs 0.372 0.075 4.937 0.000 0.372 0.356
col 0.123 0.084 1.465 0.143 0.123 0.108
gre 0.369 0.078 4.755 0.000 0.369 0.326
Variances:
Estimate Std.Err z-value P(>|z|) Std.lv Std.all
.gre 49.694 4.969 10.000 0.000 49.694 0.556
.grad 59.998 6.000 10.000 0.000 59.998 0.523
2.éæ¥å¹æããã³å ¨ä½å¹æ
model2 <- ' # direct effect grad ~ c*hs # mediator gre ~ a*hs grad ~ b*gre # indirect effect (a*b) ab := a*b # total effect total := c + (a*b) # another path gre ~ col grad ~ col '
ã©ãã«ãããããã«ã¤ããã
ç¬ç«å¤æ°Xã«ã¯c
å¾å±å¤æ°Yã«ã¯b
åªä»å¤æ°Mã«ã¯a
ã©ãã«ã¯ä»»æã§ãããéæ¥å¹æã®æ¸ãæ¹ã¯ä»¥ä¸ã
ã¾ã:=
ã¯æ¯æ°ãå®ç¾©ã¥ããããã®è¨å·ã§ããã
ab
ã¯a*b
ã¨å®ç¾©ã¥ãããã¦ããããããéæ¥å¹æã«ç¸å½ãããéæ¥å¹æã¯æ§æãã2ã¤ã®ãã¹ã®ç©ã§ããããå®ç¾©ãç©ã«ãªã£ã¦ããã\
totalã¨ããã®ã¯ç·åå¹æã§ãããc + (a*b)
ã¨å®ç¾©ã¥ãããããã¡ãªã¿ã«c
ã¯ç´æ¥å¹æã§ããã
fit2 <- sem(model = model2, data=df1) summary(fit2, standardized=TRUE)
3.å ·ä½çãªéæ¥å¹æ
model3 <- ' # direct effect grad ~ c*hs # mediator gre ~ a*hs grad ~ b*gre gre ~ f*col col ~ d*hs grad ~ e*col # indirect effect ab := a*b # hs -> tre -> grad dfb := d*f*b # hs -> col -> gre -> grad de := d*e # hs -> col -> grad # total effect total_ind := (a*b) + (d*f*b) '
fit3 <- sem(model = model3, data=df1) summary(fit3, standardized=TRUE)
çµæã
Regressions: Estimate Std.Err z-value P(>|z|) Std.lv Std.all grad ~ hs (c) 0.372 0.075 4.937 0.000 0.372 0.356 gre ~ hs (a) 0.309 0.065 4.756 0.000 0.309 0.335 grad ~ gre (b) 0.369 0.078 4.755 0.000 0.369 0.326 gre ~ col (f) 0.400 0.071 5.626 0.000 0.400 0.396 col ~ hs (d) 0.605 0.048 12.500 0.000 0.605 0.662 grad ~ col (e) 0.123 0.084 1.465 0.143 0.123 0.108 Variances: Estimate Std.Err z-value P(>|z|) Std.lv Std.all .grad 59.998 6.000 10.000 0.000 59.998 0.523 .gre 49.694 4.969 10.000 0.000 49.694 0.556 .col 49.025 4.903 10.000 0.000 49.025 0.561 Defined Parameters: Estimate Std.Err z-value P(>|z|) Std.lv Std.all ab 0.114 0.034 3.362 0.001 0.114 0.109 dfb 0.090 0.026 3.487 0.000 0.090 0.086 de 0.075 0.051 1.455 0.146 0.075 0.071 total_ind 0.204 0.047 4.332 0.000 0.204 0.195
hs -> tre -> grad: ab = 0.114
hs -> col -> gre -> grad : dfb = 0.090
hs -> col -> grad : de = 0.075
Effects from HS to GRAD via GRE
Sum of indirect : ab + dfb = 0.204
4.éå°åå®ã¢ãã«
model4 <- ' col ~ hs gre ~ col grad ~ col grad ~ gre ' fit4 <- sem(model = model4, data=df1) summary(fit4, standardized=TRUE, fit.measures=TRUE) #ãã£ããææ¨ã®è¡¨ç¤º
çµæã
Estimator ML Optimization method NLMINB Number of model parameters 7 Number of observations 200 Model Test User Model: Test statistic 44.429 Degrees of freedom 2 P-value (Chi-square) 0.000 Model Test Baseline Model: Test statistic 362.474 Degrees of freedom 6 P-value 0.000 User Model versus Baseline Model: Comparative Fit Index (CFI) 0.881 Tucker-Lewis Index (TLI) 0.643 Loglikelihood and Information Criteria: Loglikelihood user model (H0) -2062.830 Loglikelihood unrestricted model (H1) -2040.616 Akaike (AIC) 4139.660 Bayesian (BIC) 4162.748 Sample-size adjusted Bayesian (BIC) 4140.571 Root Mean Square Error of Approximation: RMSEA 0.326 90 Percent confidence interval - lower 0.247 90 Percent confidence interval - upper 0.412 P-value RMSEA <= 0.05 0.000 Standardized Root Mean Square Residual: SRMR 0.102 Parameter Estimates: Standard errors Standard Information Expected Information saturated (h1) model Structured Regressions: Estimate Std.Err z-value P(>|z|) Std.lv Std.all col ~ hs 0.605 0.048 12.500 0.000 0.605 0.662 gre ~ col 0.625 0.056 11.101 0.000 0.625 0.617 grad ~ col 0.317 0.079 4.014 0.000 0.317 0.276 gre 0.492 0.078 6.303 0.000 0.492 0.434 Variances: Estimate Std.Err z-value P(>|z|) Std.lv Std.all .col 49.025 4.903 10.000 0.000 49.025 0.561 .gre 55.313 5.531 10.000 0.000 55.313 0.619 .grad 67.311 6.731 10.000 0.000 67.311 0.587