Which analyses makes sense?

[StatsUser]

CONTENTS DELETED

The author has deleted this message.

[David Marso]

Please reread the very useful suggestions you were provided with yesterday in your previous posting!

StatsUser wrote

Dear Listserv,

I'm trying to figure out which analyses makes the most sense to do.

I have 4 dependent, continuous variables (competence baseline, competence violation, competence repair, competence redress) and 1 independent categorical variable with 2 levels (trust repair: apology, control).

IV is a between subjects variable (different participants for apology group and different participants for control group).

DV is a repeated measures variable - participants in apology group completed competence measure 4 times - baseline, violation, repair, redress, whereas a different set of participants in the control group completed their own set of competence measures at 4 points in time)

Research questions is : do percpetions of competence improve once an apology is issued?

This is an experimental design - sequence is:


Apology condition:
Read article
DV measure 1: competence baseline

Continue reading article that shows competence violation
DV measure 2: competence violation

Continue reading article that shows apology
DV measure 3: competence repair

Continue reading article that shows redress plan
DV measure 4: competence redress

Control condition: same sequence just all neutral passages

Thanks,

[Bruce Weaver]

"I have 4 dependent, continuous variables (competence baseline, competence violation, competence repair, competence redress)..."

I would describe that as ONE dependent variable (competence) measured under 4 different conditions.  I.e., you have a repeated measures situation, not a multivariate one.  So as David notes, the model that was described yesterday is applicable here too.  But one obvious difference is that your repeated measures factor now has more than 2 levels.  Therefore, lack of sphericity may be an issue.  For more information on sphericity, and how to deal with lack of it, see Thom Baguley's nice note.

  http://homepages.gold.ac.uk/aphome/spheric.html

HTH.

StatsUser wrote

Dear Listserv,

I'm trying to figure out which analyses makes the most sense to do.

I have 4 dependent, continuous variables (competence baseline, competence violation, competence repair, competence redress) and 1 independent categorical variable with 2 levels (trust repair: apology, control).

IV is a between subjects variable (different participants for apology group and different participants for control group).

DV is a repeated measures variable - participants in apology group completed competence measure 4 times - baseline, violation, repair, redress, whereas a different set of participants in the control group completed their own set of competence measures at 4 points in time)

Research questions is : do percpetions of competence improve once an apology is issued?

This is an experimental design - sequence is:


Apology condition:
Read article
DV measure 1: competence baseline

Continue reading article that shows competence violation
DV measure 2: competence violation

Continue reading article that shows apology
DV measure 3: competence repair

Continue reading article that shows redress plan
DV measure 4: competence redress

Control condition: same sequence just all neutral passages

Thanks,

[Bruce Weaver]

Here's a more nicely formatted version of Thom Baguley's note on sphericity.

   http://psychologicalstatistics.blogspot.ca/2006/05/what-is-all-this-stuff-about.html

Bruce Weaver wrote

"I have 4 dependent, continuous variables (competence baseline, competence violation, competence repair, competence redress)..."

I would describe that as ONE dependent variable (competence) measured under 4 different conditions.  I.e., you have a repeated measures situation, not a multivariate one.  So as David notes, the model that was described yesterday is applicable here too.  But one obvious difference is that your repeated measures factor now has more than 2 levels.  Therefore, lack of sphericity may be an issue.  For more information on sphericity, and how to deal with lack of it, see Thom Baguley's nice note.

  http://homepages.gold.ac.uk/aphome/spheric.html

HTH.

StatsUser wrote

Dear Listserv,

I'm trying to figure out which analyses makes the most sense to do.

I have 4 dependent, continuous variables (competence baseline, competence violation, competence repair, competence redress) and 1 independent categorical variable with 2 levels (trust repair: apology, control).

IV is a between subjects variable (different participants for apology group and different participants for control group).

DV is a repeated measures variable - participants in apology group completed competence measure 4 times - baseline, violation, repair, redress, whereas a different set of participants in the control group completed their own set of competence measures at 4 points in time)

Research questions is : do percpetions of competence improve once an apology is issued?

This is an experimental design - sequence is:


Apology condition:
Read article
DV measure 1: competence baseline

Continue reading article that shows competence violation
DV measure 2: competence violation

Continue reading article that shows apology
DV measure 3: competence repair

Continue reading article that shows redress plan
DV measure 4: competence redress

Control condition: same sequence just all neutral passages

Thanks,

[Ryan]

Are these subscales of a larger psychological test measuring competence, each of which was standardized [on the validation sample] to have the same mean and standard deviation?

 

Ryan

On Fri, Jul 27, 2012 at 8:40 AM, Bruce Weaver <[hidden email]> wrote:

"I have 4 dependent, continuous variables (competence baseline, competence

violation, competence repair, competence redress)..."

I would describe that as ONE dependent variable (competence) measured under
4 different conditions.  I.e., you have a repeated measures situation, not a
multivariate one.  So as David notes, the model that was described yesterday
is applicable here too.  But one obvious difference is that your repeated
measures factor now has more than 2 levels.  Therefore, lack of sphericity
may be an issue.  For more information on sphericity, and how to deal with
lack of it, see Thom Baguley's nice note.

  http://homepages.gold.ac.uk/aphome/spheric.html

HTH.

StatsUser wrote
>
> Dear Listserv,
>
> I'm trying to figure out which analyses makes the most sense to do.
>
> I have 4 dependent, continuous variables (competence baseline, competence
> violation, competence repair, competence redress) and 1 independent
> categorical variable with 2 levels (trust repair: apology, control).
>
> IV is a between subjects variable (different participants for apology
> group and different participants for control group).
>
> DV is a repeated measures variable - participants in apology group
> completed competence measure 4 times - baseline, violation, repair,
> redress, whereas a different set of participants in the control group
> completed their own set of competence measures at 4 points in time)
>
> Research questions is : do percpetions of competence improve once an
> apology is issued?
>
> This is an experimental design - sequence is:
>
>
> Apology condition:
> Read article
> DV measure 1: competence baseline
>
> Continue reading article that shows competence violation
> DV measure 2: competence violation
>
> Continue reading article that shows apology
> DV measure 3: competence repair
>
> Continue reading article that shows redress plan
> DV measure 4: competence redress
>
> Control condition: same sequence just all neutral passages
>
> Thanks,
>

-----
--
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/Which-analyses-makes-sense-tp5714492p5714498.html

Sent 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