Date:         Fri, 12 Dec 2008 11:37:57 -0600
Reply-To:     Mary <mlhoward@avalon.net>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Mary <mlhoward@AVALON.NET>
Subject:      Re: experimental design question
Comments: To: "Swank, Paul R" <Paul.R.Swank@UTH.TMC.EDU>
Content-Type: text/plain; charset="iso-8859-1"

Paul,

It sounds to me like it is measured; presumably at baseline and at followup
(time of assesment of improvement or not). 

I'd be interested in your thoughts on what sort of model this would be.

-Mary
  ----- Original Message ----- 
  From: Swank, Paul R 
  To: SAS-L@LISTSERV.UGA.EDU 
  Sent: Friday, December 12, 2008 11:34 AM
  Subject: Re: experimental design question


  It looks like you are positing a mediation model but it is unclear about
  X enzyme level. Is this measured or manipulated?

  Paul R. Swank, Ph.D
  Professor and Director of Research
  Children's Learning Institute
  University of Texas Health Science Center
  Houston, TX 77038


  -----Original Message-----
  From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of duo
  wan
  Sent: Friday, December 12, 2008 11:11 AM
  To: SAS-L@LISTSERV.UGA.EDU
  Subject: experimental design question

  Hi All,

  Could someone help me with a quick question for experimental design?

  The design is a simple one:
  1  New therapy has better treatment results than old therapy.
  2. high X enzyme level is associated  with  better  treatment results.

  Test hypothesis: New therapy is better probably because it produces
  higher X enzyme level.

  "New therapy=>higher X enzyme level=>better treatment results".
  One corresponding design solution:
  Group1: New therapy + nothing (higher X)
  Group2: New therapy+X enzyme inhibitor(lower X)
  Group3: Old therapy+nothing (lowest X level)

  can ANOVA answer "New therapy=>higher X enzyme level=>better treatment
  results" by comparing 3 groups in term of treatment results?

  I think it is more than ANOVA because X enzyme level is an intermediate
  outcome, not just explanatory variable. or This is question that should
  be answered by causal model (direct, indirect effects,surrogate
  outcome)?

  Anyone could help me with some references with similar design?

  thanks,

  Vincent