G**7 发帖数: 391 | 1 I am trying to run latent growth analysis. Part of the interest is to
get predicted values from LISREL. Having searched for a while, I couldn't
find the command language to do that.
Anyone knows how to get the predicted values from LISREL?
BTW, the key words I have been using with search are: predicted, LISREL
syntax.
In addiction, I also downloaded several LISREL files to search. Still
nothing. |
|
c******s 发帖数: 2163 | 2 我从来没使用过lisrel , 但是今天帮朋友的忙必须要用这个, 我们写好了CODE 目的
是create covariance matrix, 但是每次运行都会出现:fatal error: error in raw
data for case 1: 后面就是一堆0. 真不知道原数据有什么问题。
哪位给指点一下, 非常感谢。 |
|
A***A 发帖数: 98 | 3 ape同学好,前段时间忙死了。 John Horn 光看上去就是挺牛的,白胡子一大把。
说个有意思的,去年我们造那个factor analysis family tree 的时候,记不得是Karl
Joreskog (of the Lisrel fame) 还是 Michael Browne 说
这个tree上面有很多"bastards"
(原话),就好比他们自己,不是Spearman或者Thurstone的师承,而是先搞数理统计,后
对psychometrics感兴趣。
已经 |
|
h****g 发帖数: 28 | 4 thx..
我其实是要分析潜变量的调节作用
由于都是用的AMOS软件进行分析,不会用LISREL
所以现在看来只好将调节变量分组然后采用AMOS分析了
BTW.
我在文献中查阅到Joreskog-Yang 模型可以用乘积来作,
如潜变量X有俩指标X1和X2、调节变量M有俩指标M1M2,结果变量Y;
那么X1M1、X1M2、X2M1、X2M2作为调节效应潜变量XM的指标加入到模型中进行
SEM分析;如果XM对Y的作用显著则调节效应显著。
觉得这种方法确实不错,但是实在不知道其中还需要设定哪些限制,
只好用简单点的分组检验sigh
多谢提醒,我已经将数据中心化了 |
|
r******s 发帖数: 2155 | 5 AMOS is similar to Lisrel. no need to switch.
The easiest way to test moderation is using regression.
Many problems exist in testing it in SEM. |
|
y***x 发帖数: 737 | 6 同行握手哈!
SPSS 早就不够用了,现在都要求会AMOS,LISREL, Mplus,sas R, 学无止境阿
2013 |
|
|
x**g 发帖数: 807 | 8 不要用SAS。 Proc Calis不能处理Missing Data。 还是用Lisrel, Amos,或是Mplus
好。WinBugs也好。 |
|
|
g******7 发帖数: 19 | 10 one possible solution is using the simutaneous equation modeling, or
structural equation modeling (the so-called SEM)
but first you will need to establish a theoretical frame/model. if you can,
it is very straitforward to test: coefficient of Z = coefficient of W by
model comparison (something similar to GLH test in GLM models).
In SAS, you use proc Calis
the other software to do it include: LISREL, AMOS etc. |
|
|
s*******r 发帖数: 769 | 12 好像这两个做SEM是最出名的
SAS的proc calis 不是也可以做 SEM 吗?好像很少有人提及似的,大部分人都在说
AMOS, LISREL |
|
l******g 发帖数: 2429 | 13
AMOS, LISREL 正版可以做大批量variable,学生版是限制在了8-16个variable
做SEM这两个软件是最最专业的,SAS做SEM纯粹是业余 |
|
l******g 发帖数: 2429 | 14
不是说Mplus,SAS这些不好,只不过跟AMOS, LISREL比起来,做SEM太业余了点。。。。
现在也有很
多人用HLM/HMM,不过模型已经不一样了, |
|
G**7 发帖数: 391 | 15 I am trying to figure out the LISREL syntax for the latent growth model (
with both confirmatory latent factor and covariate).
If you know any related syntax and can share, I would appreciate that.
Thanks. |
|