由买买提看人间百态

boards

本页内容为未名空间相应帖子的节选和存档,一周内的贴子最多显示50字,超过一周显示500字 访问原贴
Psychology版 - 请问大家有关mixed model (转载)
相关主题
ask for help about sas contrast statement-包子酬谢 (转载)Self-fulfilling prophecy和Pygmalion effect
Baozi award: event in survival analysis (转载)panic disorder 有救吗? (转载)
Short Course on Unfolding IRT Models[合集] Forer Effect和星座学说
Help: R softwareMental Health (for educating, BM pls mark) (转载)
有对教育心理+多媒体感兴趣的吗?job opportunity
Asiancat and Psychology please!how to find a psychologist, who can communicate with you well?
to BluegrassKH1b job
Forer Effect和星座学说统计一下
相关话题的讨论汇总
话题: timepoint话题: educ话题: mixed话题: model话题: 382
进入Psychology版参与讨论
1 (共1页)
w*****1
发帖数: 473
1
【 以下文字转载自 Statistics 讨论区 】
发信人: wz99331 (dotti), 信区: Statistics
标 题: 请问大家有关mixed model
发信站: BBS 未名空间站 (Wed Jul 19 12:57:18 2017, 美东)
大家好,我需要用mixed model 对三个group的人群(教育程度1,2,3)进行分析,看
看这些人的认知能力在不同时间的变化有没有显著性的不同。
proc mixed data=outban;
class timepoint educ;
model rbantotal= timepoint | educ/ solution;
random int /subject=id ;
lsmeans educ*timepoint;
run;
因为我是3个group,而不是两个group,如果我想看相互作用项educ*timepoint的p-值,
是不是需要用f-test?如何在mixed model中用f-test呢?我看到现在的结果里就已经有
f-test了:
Type 3 Tests of Fixed Effects
Num Den
Effect DF DF F Value Pr >
F
timepoint 1 382 8.11 0.
0046
educ 3 382 1.85 0.
1381
timepoint*educ 3 382 1.78 0.
1512
但是是fixed effects,这个f-test结果就是我们要的相互作用项的p-value吗?非常感
谢!回答有帮助的朋友包子奖励。
1 (共1页)
进入Psychology版参与讨论
相关主题
统计一下有对教育心理+多媒体感兴趣的吗?
奖学金请教Asiancat and Psychology please!
Adler school of professional psychology,谁听说过?to BluegrassK
about school psychologyForer Effect和星座学说
ask for help about sas contrast statement-包子酬谢 (转载)Self-fulfilling prophecy和Pygmalion effect
Baozi award: event in survival analysis (转载)panic disorder 有救吗? (转载)
Short Course on Unfolding IRT Models[合集] Forer Effect和星座学说
Help: R softwareMental Health (for educating, BM pls mark) (转载)
相关话题的讨论汇总
话题: timepoint话题: educ话题: mixed话题: model话题: 382