- 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:% r% P- u+ o9 ~3 ^
5 W7 p1 }9 E( @1 l a- X, {5 ?6 d
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. * W. Q9 n6 c* S. |6 z7 v
/ V/ D& |+ r5 h5 E
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.
: S) ~4 p' T$ W- g, m, a4 T, u; X/ y
在处理cross-level mediation的时侯,作者3 q/ W z3 |, U4 a k5 w% y7 q8 J
“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 U' A0 c5 d0 }2 D3 |
1 w4 Y8 `2 S4 M; F6 Y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. % m3 O# u5 D9 _, r$ @
4 V. r" c. {9 Z; g( e7 D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.”! f3 O0 C6 Q2 G5 W
. T0 J; t. H% \- N1 r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)。: \2 R7 R3 [5 N
% w- f; t' a, s S8 ^
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。
{* b$ h0 y4 T" r m" C: f
+ c, C- N& V( }2 j# r. `4 ^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? & m3 _/ A6 x8 P$ N( h: u
u' L4 H, g- b+ K& T. n U
真诚希望得到您的指点!
; T X: e0 }+ j5 Z% k$ U. t; o
" q! ]+ t4 P3 A9 F/ n: a3 S, |2 g谢谢先!
+ N. A, }/ d! ]# J! A- @7 G5 a6 V$ {# S: n) w
0 y% P' g; s, C, J Y% X$ G ]* ]+ a7 K! N0 o A
\" y- l; x3 `5 E' N
; A4 H% d+ Z- r. ^6 h5 R
! c- w& v) v( P' Y# j8 k+ D! [6 f% r |
|