- 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:
& }. N" {$ x( u6 l H+ S: N; ~' Q) T& W# F( M
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.
* H* ]2 N1 @0 R) v* X- }& l
! c5 ~' E9 q2 X6 I$ b3 C! b: kNohe, 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.
& b9 F2 b' B" _; K
) h+ J) E# ]: C& c3 h在处理cross-level mediation的时侯,作者
% [( i, B9 T: m$ i3 V& m“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]
v+ j, C9 M& i* n y: I. x5 S/ R& W; F9 S7 R
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. 2 z- J1 C; ?# u# s; Q
) c; ]' L4 \# K6 K/ R" ?5 V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.”5 l: k: h3 m; z* c6 {
4 J, x# w: s5 ^( 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)。( L1 B, T5 N' f* }: J* b2 P3 d
, T$ h* k+ b$ l" w/ ^- O* V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。 0 ?$ B3 |. k* _ |5 q
: q1 p* J4 H& K1 |
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? # y+ i o! B9 S2 U4 W; e5 r L
7 u3 s: @4 V5 [# Y7 v# S* M真诚希望得到您的指点!7 p/ p/ P1 t; ~- u
# m' Y4 R$ C, w4 W$ o谢谢先!
4 t2 F. j# Y, k/ p) y1 _' q
( e, n% y0 P4 ^. O& G/ T7 C* ^! J) W( e: E& Y, h- N. y
- ]' f6 ?* }! m) k- \! a' H
, m: |3 i, s' y |! ?1 z9 U* y- i) n8 e
, V! M; O- q- {" g/ O G |
|