Date:         Tue, 14 Jan 2003 15:47:41 -0500
Reply-To:     "Burleson,Joseph A." <burleson@up.uchc.edu>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Burleson,Joseph A." <burleson@up.uchc.edu>
Subject:      Re: Repeated measures
Comments: To: Bautista Christian <cbautista@NMRCD.MED.NAVY.MIL>
Content-Type: text/plain; charset="iso-8859-1"

Use GLM. One between-subjects variable (Treatment), one within-subjects
variable (time). Look at the Time X Treatment interaction. If it is sig, one
group improved more than the other.

-Joe Burleson

-----Original Message-----
From: Bautista Christian [mailto:cbautista@NMRCD.MED.NAVY.MIL]
Sent: Tuesday, January 14, 2003 8:19 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Repeated measures


Hi list,

I have a trial study where there are two groups (placebo and intervention).

STEP 1. At the beginning of the study I have collected a quantitative
measure (weight) for each participant in each group.

STEP 2. Then, we have given a special diet for all participants of the
intervention group and a normal diet for the place group.

STEP 3. After 10 days we have measured their weigth again for all
participant in each group.

ANALYSIS PLAN
a) Apply paired-samples t-test to evaluate the difference OF WEIGHT
(before-after) in each group separately.

QUESTION
I need to know if there is a difference of weight between both groups.

What statistical test I have to use?

I was thinking in General Lineal Model for repeated measures but I have only
two measures.

PROBABLE SOLUTION
I migth calculate the difference of weight (before-after) for each group and
then to evaluate this difference using a simple Student t test.

Thanks in advance,

/Christian