Hi Bruce,
I read the first part of your message, where you described how to fill in the menu fields and which answered the posted question, but I didn't scroll down to read the rest of your message. My mistake. Functionally, there is no difference that I know of between using EMMEANS and using either Preacher's website or Hayes' macro. Like you said, when using EMMEANS, he will need to cycle through values of the covariate to identify point (or points) where the region(s) of significance begin. That will have to be done twice by twice, twice for TX group 1 and twice for TX group 2 because he will need to establish whether there is an upper, lower, or both region of significance for each TX group value. Those iterations is hidden in website and macro.
Gene Maguin
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of Bruce Weaver
Sent: Thursday, September 08, 2016 12:22 PM
To:
[hidden email]Subject: Re: ANCOVA SPSS Question HELP
Hi Gene. Just out of curiosity, why would the macros you refer too add to what you get from the EMMEANS sub-commands I included? They show the pair-wise comparisons among the 3 groups at selected levels of the moderator variable. (You weren't thinking about mediation, by any chance, were you?
That's a completely different kettle of fish.)
Cheers,
Bruce
Maguin, Eugene wrote
> (Others) Credibility established. But, wasn't this a more complex
> homework problem than would be expected on first or second week of the semester?
>
> To add to what Bruce posted.
>
> If the TX*PET term (PET = pre-existing trait) is significant, you'll
> need to look at a moderation analysis, which has an established
> literature (and a downloadable spss macro). One key name is Kristopher
> Preacher (quantpsy.org); another is Andrew Hayes.
> Gene Maguin
>
>
>
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:
> SPSSX-L@.UGA
> ] On Behalf Of James Ross
> Sent: Wednesday, September 07, 2016 5:13 PM
> To:
> SPSSX-L@.UGA
> Subject: Re: ANCOVA SPSS Question HELP
>
> I should give more info since I believe my first message was
> interpreted as someone who is doing homework for a college or high
> school class. Here are the specifics of what I need help with:
>
> I gathered data on 1000+ participants and measured their attitudes via
> a standardized measure that uses Likert scale responses. This measure
> which I'll call my attitude measure was given before and after an intervention.
>
> There were three groups for the intervention. Participants were
> randomly assigned to Intervention Group A, Intervention Group B, and a
> control group. The control group was not given any intervention but
> did a random task. All three groups completed the Pre-Test and
> Posttest attitude measure. I am hypothesizing that Groups A's
> intervention post-test scores will be less than Group B's intervention
> post-test scores and that both A and B group's post-test scores will
> be less than the control group's post-test score.
>
> In addition, before administering my attitude measure, I administered
> a scale that is meant to measure a specific pre-existing trait. This
> measure also uses Likert scale responses and produces an ordinal sum
> total ranging from 10 to 50. I'm calling this my moderating variable
> because I believe that the Pre and Post test scores will decrease even
> more for those that have a higher score on my trait measure. But I
> don't understand how to put all these scores in SPSS.
>
> Someone earlier said that this was not an ANCOVA, but I thought any
> pre-post test comparisons with group assignments had to have the
> Pre-test scores put down as a Covariate? Plus I don't know where to
> put (what I believe is moderating) my trait measure score?
> Conceptually, isn't it also a covariate or is it something else?
>
> I hope this clears up earlier message.
> Thanks,
> James
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> LISTSERV@.UGA
> (not to SPSSX-L), with no body text except the command. To leave the
> list, send the command SIGNOFF SPSSX-L For a list of commands to
> manage subscriptions, send the command INFO REFCARD
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> LISTSERV@.UGA
> (not to SPSSX-L), with no body text except the command. To leave the
> list, send the command SIGNOFF SPSSX-L For a list of commands to
> manage subscriptions, send the command INFO REFCARD
-----
--
Bruce Weaver
[hidden email]http://sites.google.com/a/lakeheadu.ca/bweaver/"When all else fails, RTFM."
NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/ANCOVA-SPSS-Question-HELP-tp5733064p5733082.htmlSent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD