- 最后登录
- 2014-9-15
- 注册时间
- 2013-11-14
- 威望
- 0
- 金钱
- 65
- 贡献
- 40
- 阅读权限
- 20
- 积分
- 105
- 日志
- 0
- 记录
- 0
- 帖子
- 17
- 主题
- 4
- 精华
- 0
- 好友
- 0
  
- 注册时间
- 2013-11-14
- 最后登录
- 2014-9-15
- 积分
- 105
- 精华
- 0
- 主题
- 4
- 帖子
- 17
|
罗老师好,想请教您一个问题:
8 i' v& p: _. j9 q+ t7 d m: P' _! }1 P) h
我在Leadership Quaterly上面看到一篇文章,是用MSEM测试的 一个2-1-1-2 sequential mediation.
0 l8 o9 i# k0 a% T! K: R8 |/ J3 C7 k. Q1 \# C9 h
Nohe, C., Michaelis, B., Menges, J. I., Zhang, Z., & Sonntag, K. (2013). Charisma and organizational change: A multilevel study of perceived charisma, commitment to change, and team performance. The Leadership Quarterly, 24, 378-389. b8 E0 p1 t* N* I# ?/ ^
' U$ r& G; a2 a2 a& A9 b) V% x在处理cross-level mediation的时侯,作者' w- ?. J) v; R* G' x
“decompose the variance of a variable into its latent within-unit component (within-team variance) and a latent between-unit component (between-team variance; Lüdtke et al., 2008; Muthén & Asparouhov, 2009). Relationships between these variance components can be modeled independently at each level through the specification of structural and measurement models.”[p.383]( _) y! i8 z* f0 U1 B
# E4 d/ B! }/ |4 U按照这个方法,我用Mplus把level 1的variable 分别在Level 1和Level 2作为latent variable 建Model (两个level 1的variable,每个都由7个item来衡量),然后只用Level 2的coefficients来计算the indirect effect of sequential mediation.
! ]) k: a9 y3 z" |; ?
2 N* c7 C4 d% U. b$ U虽然Mplus给出了coefficient的estimation和significance level,但是model整体“nonidentified”,然后Mplus report “THE NONIDENTIFICATION IS MOST LIKELY DUE TO HAVING MORE PARAMETERS THAN THE NUMBER OF CLUSTERS. REDUCE THE NUMBER OF PARAMETERS.”
! b: M6 ^' s, N% u% X2 }5 P& m
# U+ x& P w$ S8 M我想这大概是主要因为每个latent variable的measure item比较多,可是group number只有39个的原因(虽然individual level sample 有1800多个)。
6 x6 q* C/ w! l. c- e H
& g) h, k' \9 {7 f5 o我试了另外一个model,不是用latent variable,而是用直接的aggregated variable分别在Level 1(mean of the 7 measure items) 和Level 2 (group average of the level 1 variable)参与模型运算。还是只看所有Level 2的coefficients来计算2-1-1-2的indirect sequential mediation。 Mplus就没有任何warning了。 . G0 t; g1 ^! c- F/ w
" E& N# ]# H' C8 n% T W我想请问罗老师,如果我在limitation里面承认model有不完美的地方,我可以用我自己试的方法,不用latent variable,而是直接在Level 1和Level 2用aggregated value来计算一个2-1-1-2的模型么? * ^ J' z0 }" R1 U7 N# V3 t
; y9 Q2 K; S" @1 k& k$ y9 C
真诚希望得到您的指点!
3 P3 z9 B* v2 ]
8 T) y6 P% [6 |4 }$ Z) }! @谢谢先!
6 R* i @ V. L0 ~
) X# L9 g' j2 e; }' \2 B; Q& A% }
! ?. n3 L7 L& v, B k9 L5 ?# s& o4 f4 {# ^! f$ f4 @
% O9 D/ I8 p4 W# q7 z5 K! o% H) ~6 s3 Y- T: U
|
|