Date: Sat, 23 Mar 2002 15:23:58 -0300
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Hector Maletta <email@example.com>
Subject: Re: A Really Stupid Question . . .
Content-Type: text/plain; charset=us-ascii
Geraldine, just be literal: ""reference category =2" means the
reference category is the one which is coded 2, i.e. "Low" in Factor A
and "Absent" in Factor B. The effect of a High level of Factor A, and
the effect of Factor B being Present is measured as a difference against
the effect of A=Low and the effect of B absent, respectivaly. Comparing
level 1 to level 2 means comparing the group coded 1 to the group coded
2 for whichever factor the phrase is applied to.
Universidad del Salvador
Buenos Aires, Argentina
"Geraldine R. Henderson" wrote:
> Okay, I have been away from Statistical Analysis for a while, back when
> we ran SAS on a mainframe. Anyway, I am trying to figure out how SPSS
> (10.1 for Windows) operates. I am trying to run planned simple
> contrasts in MANOVA on two factors that had a significant interaction.
> Each of the two factors has only two levels. Factor A is coded such
> that the true value "high" is coded as "1" and the true value "low" is
> coded as "2." Likewise, Factor B is coded such that the true value
> "present" is coded as "1" and absent is coded as "2."
> So here is my really stupid question:
> When my SPSS simple contrast on Factor A output tells me the the
> "reference category =2" which I assume is the last category, how do I
> know if the reference category is "high" or "low?" In addition, when it
> tells me that it is comparing "Level 1" vs. "Level 2," which is which?
> "Present" or "Absent?" How does it assign which level is which? Is it
> based on order of data entry, ordinal value even for nominal data?
> I have tried books, calling SPSS (Statistician will call me back by EOD
> Monday), AnswerNet, etc. I know that this is a really stupid
> question, but I can't seem to find the answer.
> Will someone out there help me once you stop laughing at the idiocy of
> my message?
> Thanks, Gerri.