Which test to use?

#1
This is my experiment.

I have 3 independent variables.
1. sex of participant
2. sex of stimuli (sex of pictorial stimuli in the experimental task)
3. level of anxiety

I have 2 dependent variables.
1. bias score (Response time)
2. accuracy (accuracy of responses in experimental task)

I would like to see the interactions and main effects of these variables, but I'm not sure which test to use.
Also, should I code the level of anxiety into two - anxious and nonanxious- using a cutoff score? Or can I just leave it as a continuous variable?

Data is normal and sample sizes are as below.
1. male anxious - 15
2. male nonanxious - 15
3. female anxious - 15
4. female nonanxious - 15
TOTAL = 60

Thank you in advance.
 

Karabiner

TS Contributor
#2
This is my experiment.

I have 3 independent variables.
1. sex of participant
2. sex of stimuli (sex of pictorial stimuli in the experimental task)
3. level of anxiety
Seemingly, sex of stimulus is a within-subject (repeated-measures) factor?
If this is (in part) a repeated-measures design, then how many stimuli did the subjects see?

Also, should I code the level of anxiety into two - anxious and nonanxious- using a cutoff score? Or can I just leave it as a continuous variable?
Never dichotomize contiuous variables, except if there is a very strong reason for that.
It destroys information and is against common sense.
http://www.psychology.sunysb.edu/attachment/measures/content/maccallum_on_dichotomizing.pdf
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1458573/
http://biostat.mc.vanderbilt.edu/wiki/Main/CatContinuous

You can use the continuous measurement.

Data is normal
Reaction times = normal? Never ever.

If you have a repeated measures design, you could consider repeated-measures analysis
of variance, with the additional factor "sex of participant" and the additonal continuous
predictor "anxietey score". "Sex of stimulus" seemingly is a repeated-measures (within-
subject) factor.

Maybe you want to recode stimulus with respect to sex of the participant (same-sex
stimulus vs. opposite-sex stimulus)?

With kind regards

Karabiner
 
Last edited:
#3
Thank you so much for your help!! :) I really appreciate it. Really!!

Seemingly, sex of stimulus is a within-subject (repeated-measures) factor?
If this is (in art) a repeated-measures design, then how many stimuli did the subjects see?
The subjects saw two types of stimuli- male, and female.

I will try recoding as you suggested.

Thank you! :)