结构方程论坛SEM-Structural·Equation·Modeling

标题: 悬赏50论坛币问下我这种情况怎么做AMOS分析 [打印本页]

作者: 华南心理学 时间: 2011-9-22 00:21 标题: 悬赏50论坛币问下我这种情况怎么做AMOS分析

我想做一个行业的投入与产出的相关关系分析研究现有的数据是这个行业各个企业的投入和产出方面的数据请问适合用AMOS做研究吗？这些数据如何处理如果不适合需用什么软件？做信度检验时，投入和产出方面的信度都很低，而且有些产出指标的潜变量只能有1个指标组成，可以用AMOS嘛，怎么办谢谢各位了

作者: sopheia 时间: 2011-9-22 00:26

First in AMOS can use as data input a correlation matrix . Here is a link (download the .sps syntax file) http://core.ecu.edu/psyc/wuenschk/MV/SEM/Path-SPSS-AMOS.doc and, http://groups.google.com/group/c ... 9f08cf7766d2b?pli=1 Second, if your reliability is low, usually<0.6 that means the item correlations are also low. That is you can see in SPSS (check all the descriptives under the statistics button) in reliability dialog. So, you can drop the problematic variables. Furthermore, you must see if any of your items needs reverse coding according to your questinnaire. After that when your reliability is acceptable, you can import your correlation matrix. Third there are some rules how many indicators are in any latent variable. The next is from Kline's book(page 138), "

RINCIPLES AND PRACTICE OF STRuCTuRAL EQuATION MODELING, Third Edition Rex B. Kline". If a standard CFA model with a single factor has at least three (Rule 6.4) indicators, the model is identified. If a standard CFA model with ≥ 2 factors has ≥ 2 indicators per (Rule 6.5) factor, the model is identified. I wish these are helps you

欢迎光临结构方程论坛SEM-Structural·Equation·Modeling (http://semchina.net/)