- 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:
! w6 X' L- n; i) \2 @- u# q
9 Q `) r/ P6 r6 |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. 1 ~! m1 h d( @& T" ]
( u# C- I; v& a! j: ~: B4 E) RNohe, 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, 378-389.9 T* [* m; Q- o: V w
9 t3 @& ]1 S* |' K, f在处理cross-level mediation的时侯,作者! Y; u/ a. o" j$ b
“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]1 H. ~' W" E. r! u* l
) q* q- i7 ~: q$ c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. + ~. m( |8 ~6 g! Q* S7 |+ p
$ j$ B2 t/ Y3 l
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.”
: t' L* h4 A3 N2 T+ I5 v# G
9 j3 T* y8 J( ?9 W* U5 W! P$ K Z) J9 x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)。7 d Y; p" b8 V% D
* q+ l% [& j& U% Q+ ~9 t1 x
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。
* o; A7 a0 {+ M+ x h% m6 N+ z: s' o
) r ]- o% F) ]" m$ ~$ E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? ) C6 A0 M$ {/ W5 T6 L* @" w. Q9 Y6 P
2 k3 ` t/ a* f/ X; s$ n真诚希望得到您的指点!8 k( P g; J x6 U( G* p
) W" n$ T7 D. t5 j- P6 l
谢谢先!8 y) j3 ]5 \ g. ^, _. d' s8 D! j
9 A1 x+ x8 M# f I
# f2 Z$ E6 g( n; B. Z% ]6 g+ P
* }% u7 e7 \) Q, J/ e1 k1 D2 [' Y. l! K3 m
% u/ z. `5 {) ]
/ E, l5 I1 Y: m- x! z' | |
|