[Code]Multilevel Modeling using Mplus

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

[size=0.8em]Level 1
    MATHij= β0j+ rij
Level 2
    β0j = γ00 + u0j

1. title:
   
2.   Introducing Multilevel Modeling by Kreft and de Leeuw.
   
3.   Page 64, Table 4.2
   
4. data:
   
5.   file = imm23.dat ;
   
6. variable:
   
7.   names = schid stuid ses meanses homework white parented public
   
8.           ratio percmin math sex race sctype cstr scsize urban region;
   
9.   cluster = schid;
   
10.   usevar  = math;
    
11.   within  = ;       ! level 1 variables here (none)
    
12.   between = ;       ! level 2 variables here (none)
    
13. 
    
14. analysis:       
    
15.   type = twolevel random;
    
16.   estimator = ml;
    
17. model:
    
18.   %within%
    
19.     math;      ! no fixed effects
    
20. 
    
21.   %between%
    
22.     math;      ! no predictors of intercept

复制代码

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏1 回帖

关键词：Multilevel Modeling Mplus Using Level

1. Level 1
   
2.     MATHij= β0j + β1(HOMEWORK) + rij
   
3. Level 2
   
4.     β0j = γ00 + u0j
   
5.     β1  = γ10
   
6. 
   
7. Here is the Mplus setup for estimating this model.
   
8. 
   
9. title:
   
10.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
11.   Page 65, Table 4.3
    
12. data:
    
13.   file = imm23.dat ;
    
14. variable:
    
15.   names = schid stuid ses meanses homework white parented public
    
16.           ratio percmin math sex race sctype cstr scsize urban region;
    
17.   cluster = schid;
    
18.   usevar  = math homework;
    
19.   within  = homework; ! level 1 variables here
    
20.   between = ;         ! level 2 variables here (none)
    
21. analysis:       
    
22.   type = twolevel random;
    
23.   estimator = ml;
    
24. model:
    
25.   %within%
    
26.     math on homework; ! fixed effect
    
27.   %between%
    
28.     math ;            ! no predictors of intercept

复制代码

1. Level 1
   
2.     MATHij= β0j + β1(HOMEWORK) + rij
   
3. Level 2
   
4.     β0j = γ00 + u0j
   
5.     β1  = γ10 + u1j
   
6. 
   
7. Here is the Mplus setup for estimating this model.
   
8. 
   
9. title:
   
10.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
11.   Page 67, Table 4.4
    
12. data:
    
13.   file = imm23.dat ;
    
14. variable:
    
15.   names = schid stuid ses meanses homework white parented public
    
16.           ratio percmin math sex race sctype cstr scsize urban region;
    
17.   cluster = schid;
    
18.   usevar  = math homework;
    
19.   within  = homework; ! level 1 variables here
    
20.   between = ;         ! level 2 variables here (none)
    
21. 
    
22. analysis:       
    
23.   type = twolevel random;
    
24.   estimator = ml;
    
25. model:
    
26.   %within%
    
27.     math ;                  ! no fixed effects
    
28.     b1 | math on homework;  ! random slope for homework
    
29. 
    
30.   %between%
    
31.     math;                   ! nothing predicts intercept
    
32.     b1;                     ! nothing predicts slope
    
33. 
    
34.     math with b1;           ! covariance between intercept and slope

复制代码

1. This model is...
   
2. 
   
3. Level 1
   
4.     MATHij= β0j + β1(HOMEWORK)  + β2(PARENTED) + rij
   
5. Level 2
   
6.     β0j = γ00 + u0j
   
7.     β1j  = γ10 + u1j
   
8.     β2  = γ20
   
9. 
   
10. Here is the Mplus setup for estimating this model.
    
11. 
    
12. title:
    
13.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
14.   Page 69, Table 4.5
    
15. data:
    
16.   file = imm23.dat ;
    
17. variable:
    
18.   names = schid stuid ses meanses homework white parented public
    
19.           ratio percmin math sex race sctype cstr scsize urban region;
    
20.   cluster = schid;
    
21.   usevar  = math homework parented;
    
22.   within  = homework parented; ! level 1 variables here
    
23.   between = ;                  ! level 2 variables here
    
24. analysis:       
    
25.   type = twolevel random;
    
26.   estimator = ml;
    
27. model:
    
28. 
    
29.   %within%
    
30.     math on parented;       ! fixed effect
    
31.     b1 | math on homework;  ! random effect
    
32. 
    
33.   %between%
    
34.     math;                   ! nothing predicts intercept
    
35.     b1;                     ! nothing predicts slope
    
36. 
    
37.     math with b1;           ! covariance intercept and slope

复制代码

1. This model is...
   
2. 
   
3. Level 1
   
4.     MATHij= β0j + β1(HOMEWORK)  + rij
   
5. Level 2
   
6.     β0j = γ00 + γ01(SCSIZE) + u0j
   
7.     β1j  = γ10 + u1j
   
8. 
   
9. The Mplus setup is shown below
   
10. 
    
11. title:
    
12.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
13.   Page 74, Table 4.7
    
14. data:
    
15.   file = imm23.dat ;
    
16. variable:
    
17.   names = schid stuid ses meanses homework white parented public
    
18.           ratio percmin math sex race sctype cstr scsize urban region;
    
19.   cluster = schid;
    
20.   usevar  = math homework scsize;
    
21.   within  = homework;         ! level 1 variables here
    
22.   between = scsize;           ! level 2 variables here
    
23. analysis:       
    
24.   type = twolevel random;
    
25.   estimator = ml;
    
26. model:
    
27.   %within%
    
28.     math;                     ! no fixed effects
    
29.     b1 | math on homework;    ! random effect of homework
    
30.   %between%
    
31.     math on scsize;           ! scsize predicts intercept
    
32.     b1;                       ! nothing predicts homework slope
    
33. 
    
34.     math with b1;             ! covariance intercept and slope

复制代码

1. This model is...
   
2. 
   
3. Level 1
   
4.     MATHij= β0j + β1(HOMEWORK)  + rij
   
5. Level 2
   
6.     β0j = γ00 + γ01(PUBLIC) + u0j
   
7.     β1j  = γ10 + u1j
   
8. 
   
9. The Mplus setup is shown below
   
10. 
    
11. title:
    
12.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
13.   Page 75, Table 4.8
    
14. data:
    
15.   file = imm23.dat ;
    
16. variable:
    
17.   names = schid stuid ses meanses homework white parented public
    
18.           ratio percmin math sex race sctype cstr scsize urban region;
    
19.   cluster = schid;
    
20.   usevar  = math homework public;
    
21.   within  = homework;        ! level 1 variables here
    
22.   between = public;          ! level 2 variables here
    
23. analysis:       
    
24.   type = twolevel random;
    
25.   estimator = ml;
    
26. model:
    
27.   %within%
    
28.     math;                    ! no fixed effects
    
29.     b1 | math on homework;   ! random effect of homework
    
30. 
    
31.   %between%
    
32.     math on public;          ! public predicts intercept
    
33.     b1;                      ! nothing predicts homework slope
    
34. 
    
35.     math with b1;            ! covariance intercept and slope

复制代码

1. This model is...
   
2. 
   
3. Level 1
   
4.     MATHij= β0j + β1(HOMEWORK)  + rij
   
5. Level 2
   
6.     β0j =  γ00 + γ01(PUBLIC) + u0j
   
7.     β1j  = γ10 + γ11(PUBLIC) + u1j
   
8. 
   
9. The Mplus setup is shown below
   
10. title:
    
11.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
12.   Page 77, Table 4.10
    
13. data:
    
14.   file = imm23.dat ;
    
15. variable:
    
16.   names = schid stuid ses meanses homework white parented public
    
17.           ratio percmin math sex race sctype cstr scsize urban region;
    
18.   cluster = schid;
    
19.   usevar  = math homework public;
    
20.   within  = homework;    ! level 1 variables here
    
21.   between = public;      ! level 2 variables here
    
22. analysis:       
    
23.   type = twolevel random;
    
24.   estimator = ml;
    
25. model:
    
26.   %within%
    
27.     math;                   ! no fixed effects
    
28.     b1 | math on homework;  ! random slope for homework
    
29.   %between%
    
30.     math on public;         ! intercept predicted by public
    
31.     b1 on public;           ! slope predicted by public
    
32. 
    
33.     math with b1;           ! covariance of intercept and slope

复制代码

1. Table 4.12 on page 82.
   
2. 
   
3. Input: ch4_p82.inp
   
4. Output: ch4_p82.out
   
5. This model is...
   
6. 
   
7. Level 1
   
8.     MATHij= β0j + β1(HOMEWORK) + β2(WHITE)  + rij
   
9. Level 2
   
10.     β0j =  γ00 + γ01(PUBLIC) + u0j
    
11.     β1j  = γ10 + u1j
    
12.     β2  = γ20
    
13. 
    
14. The Mplus setup is shown below
    
15. 
    
16. title:
    
17.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
18.   Page 82, Table 4.12
    
19. data:
    
20.   file = imm23.dat ;
    
21. variable:
    
22.   names = schid stuid ses meanses homework white parented public
    
23.           ratio percmin math sex race sctype cstr scsize urban region;
    
24.   cluster = schid;
    
25.   usevar  = math homework white public ;
    
26.   within  = homework white; ! level 1 variables here
    
27.   between = public;         ! level 2 variables here
    
28. analysis:       
    
29.   type = twolevel random;
    
30.   estimator = ml;
    
31. model:
    
32.   %within%
    
33.     math on white;          ! fixed effect of white
    
34.     b1 | math on homework;  ! random effect of homework
    
35. 
    
36.   %between%
    
37.     math on public;         ! public predicts intercept
    
38.     b1;                     ! nothing predicts homework slope
    
39. 
    
40.     math with b1;           ! covariance intercept and slope

复制代码

1. Table 4.13 on page 83.
   
2. 
   
3. Variable WHITE is now made a random effect.
   
4. 
   
5. Input: ch4_p83.inp
   
6. Output: ch4_p83.out
   
7. This model is...
   
8. 
   
9. Level 1
   
10.     MATHij= β0j + β1(HOMEWORK) + β2(WHITE)  + rij
    
11. Level 2
    
12.     β0j  =  γ00 + γ01(PUBLIC) + u0j
    
13.     β1j  = γ10 + u1j
    
14.     β2j  = γ20 + u2j
    
15. 
    
16. The Mplus setup is shown below
    
17. 
    
18. title:
    
19.   Introducing Multilevel Modeling by Kreft and de Leeuw.
    
20.   Page 83, Table 4.13
    
21. data:
    
22.   file = imm23.dat ;
    
23. variable:
    
24.   names = schid stuid ses meanses homework white parented public
    
25.           ratio percmin math sex race sctype cstr scsize urban region;
    
26.   cluster = schid;
    
27.   usevar  = math homework white public;
    
28.   within  = homework white;   ! level 1 variables here
    
29.   between = public;           ! level 2 variables here
    
30. analysis:       
    
31.   type = twolevel random;
    
32.   estimator = ml;
    
33. model:
    
34.   %within%
    
35.     math;                    ! no fixed effects
    
36.     b1 | math on homework;   ! random effect homework predicting math
    
37.     b2 | math on white;      ! random effect white    predicting math
    
38. 
    
39.   %between%
    
40.     math on public;          ! public predicts intercept
    
41.     b1;                      ! nothing predicts b1 (homework slope)
    
42.     b2;                      ! nothing predicts b2 (white slope)
    
43. 
    
44.     math with b1;            ! covariance intercept and b1
    
45.     math with b2;            ! covariance intercept and b2
    
46.     b1 with b2;              ! covariance b1 and b2
    
47. and some of the output is shown below.

复制代码