One-way ANOVA and planned comparisons - HELP!

[jwren]

Hi all,

I'm currently running a Psychology experiment: the effect of background music on memory and mood.

I will be using SPSS to conduct two one-way ANOVAs (music + mood / music + memory), w/ planned comparisons. However, I also want to see the effect of mood on memory but I don't know how to do this? (E.g. correlation?)

Any help would be much appreciated!
J

 

[Karabiner]

Could you please describe your study (research question, how large is your sample size,
what were your factors/experimental manipulations, what were your dependent variables)?

With kind regards

K.

 

[jwren]

Hi K,

Thank you for your post.
This study is based on Bottiroli et al. (2014).
We are investigating:
1) The effect of background music on cognitive performance
2) The effect of background music on mood (that is, whether 'upbeat' music elicits a more positive mood - we hypothesize this will be the case)
3) We then want to see whether mood correlates to cognitive performance (whether a positive mood, as elicited by upbeat music, leads to better cognitive performance).

This study will utilise a within-subjects design, and we are counterbalancing using the 'latin square' technique. We have 30 participants.

Independent variable (IV): Type of music (3 levels)
- Upbeat Classical - Mozart
- Downbeat Classical - Mahler
- Silence (control)

Dependant variables (DV):
- Mood after music (as determined by a mood questionnaire)
- Memory recall (as determined by a the number of words they can recall from a word list)

Hope this helps,
J.

 

[Karabiner]

3) We then want to see whether mood correlates to cognitive performance (whether a positive mood, as elicited by upbeat music, leads to better cognitive performance).

So you actually want to correlate performance with positive
mood only within the "upbeat" condition? If yes, a simple
regression would do the job, AFAICS. Or do you want to correlate
mood with performance across all 3 conditions? Or, do you even
want to analyse whether the effect of music on performance is
mediated by induced mood?

With kind regards

K.

 

[jwren]

do you want to correlate
mood with performance across all 3 conditions? Or, do you even
want to analyse whether the effect of music on performance is
mediated by induced mood?

This is where I am stuck on what to do, and unsure whether to correlate it or run further analysis.

I am inclined to go with the latter but don't know what statistical test would be used having not run an experiment like this before.

May I ask for your advice/opinion?

BW,
J.

 

[Karabiner]

I cannot tell you what you should want to find out.
If you want to perform a mediation analysis, you could
do a little reading about the concept(s) http://davidakenny.net/cm/mediate.htm ,
and there's a macro for SPSS and SAS, for the execution
of mediation analyses http://www.afhayes.com/introduction-to-mediation-moderation-and-conditional-process-analysis.html

With kind regards

K.