When to use "With" & When to Use"By"

1778

收藏 2014-05-19

I am using a random intercept model, and I have a question regarding SPSS
syntax. I would like see how a covariate is related to the dependent
variable, however, also control for the covariate effects on the
participants in the model.

To do so, in SPSS syntax, can I enter covariates following the “BY”
function and also add covariates in the FIXED function? By using this
formula, am I controlling for these covariates or should I add the
covariates using the WITH function and exclude them from FIXED section?

MIXED depression score BY Time(IV) Suicidal ideation (control) Anxiety
(control)

  /FIXED=Time(IV) Suicidal ideation (control) Anxiety (control) | SSTYPE(3)

  /METHOD=ML

  /PRINT=G  SOLUTION

/RANDOM=INTERCEPT|SUBJECT(consumid) COVTYPE(UN)

  /SAVE=PRED RESID.

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

全部回复

Lisrelchen

2014-5-19 22:04:02

First, if it is to be in the model at all it must be in the "Mixed" command. In the Mixed command use of "by" and "with" depends on the scale of measurement of the predictor. For a continuous predictor (or control) use "with" use "by" for categorical variables. Both with and by can be in the same line.
For example
MIXED depression score BY Time(IV) Suicidal ideation WITH Anxiety

If I understand the second half of your question correctly you also need to add the variable into the fixed effect. Your analysis will then give the independent contributions of your IV and of the control variable to the outcome.

扫码加我拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

扫码加我 拉你入群

扫码加我 拉你入群

分享

扫码加好友，拉您进群

扫码加我拉你入群

扫码加我拉你入群