- 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:, R0 U9 L& y. Z, {' ^4 M
/ C+ K" \& S& w; r: }* z' N6 V4 r
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. 7 o ~! [$ j+ t2 W3 A
N1 ~) u" ~: mNohe, 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, 378-389.2 C0 p3 H3 C0 Q) Z
: l: K& I1 [/ `# ]
在处理cross-level mediation的时侯,作者) d. ^9 l' x2 b y: z3 E! `
“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]
o7 A) Q; h. c, P0 U' g2 `9 d
- N" f" J& V, j0 I0 g: d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. 6 f' I w$ j, W
! c& |% t5 r% B" ~. N: M5 F4 s' ^: ]% W
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.”
, V$ I+ p* W* [/ G
9 `0 ]: J; B, `& |+ N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)。5 h+ X: ~3 u4 d& P
% R) N. {, B% E" V( a3 S1 Q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。 ! Q$ ?' g; u9 _" g5 ]
3 p$ ]5 G; v# Y, Y& s' x r
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? h4 }4 `0 f b( y* H& |! J
" H4 o0 @. F) g2 A9 n9 d6 \+ b真诚希望得到您的指点!2 b: H5 O X0 P9 E0 E- U" p
8 o( g7 m$ F2 e; s谢谢先!
' S+ w. L, d+ `. K# N! E. V/ G$ ~4 T4 w# q$ w* D% T# ]
^: x( P3 G) x6 A* F/ _
8 e' {* E2 f1 q! a7 ?
$ r7 O7 ?% y6 _ q1 w1 X, B& Y% d6 u( b4 v* i* U: r
# \* A6 M) S" v/ C( V2 d0 N& X |
|