## Best way to input and analysis this data?

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

wintersoldier
Posts: 2
Joined: Sun Mar 08, 2015 5:24 pm

### Best way to input and analysis this data?

I've got a four level IV, and a between-subjects design. Each participant was sorted into a group, and each group had its own set of stimulus. However, each group contained stimuli from each condition.

The fact that each group had content from each condition is what's confusing me. Do I just run a factorial ANOVA of each condition, or do I need to figure out a different way to code it?

Also, I'd like to display the mean search time of each stimuli per condition as a bar graph, however SPSS won't let me, is there a way around this?

Thanks.
iopsych
Posts: 25
Joined: Tue Feb 24, 2015 11:58 pm

### Re: Best way to input and analysis this data?

Can you be more specific about your design? What were the conditions, and what do you mean by "set of stimulus"? For example, did your groups experience the exact same stimuli, but in different orders?

Yalcin
wintersoldier
Posts: 2
Joined: Sun Mar 08, 2015 5:24 pm

### Re: Best way to input and analysis this data?

iopsych wrote:Can you be more specific about your design? What were the conditions, and what do you mean by "set of stimulus"? For example, did your groups experience the exact same stimuli, but in different orders?

Yalcin

Yeah sure. I think I got the design wrong; I think it's actually a within subjects design.

Conditions were: a) one symbol/one colour, b) two symbols/one colour, c) two symbols/two colours [green triangle], d) two symbols/two colours [green circle].

All participants were exposed to 12 pictures; three of each condition. However, participants were randomly sorted into 4 different groups, and thus each group so a different set of pictures (e.g. there was a total of 48 pictures, and each group saw 12 pictures).
iopsych
Posts: 25
Joined: Tue Feb 24, 2015 11:58 pm

### Re: Best way to input and analysis this data?

Hey,

I think you have a mixed design such that group is your between subjects variable and condition is your within subjects variable. You need to use the General Linear Model - Repeated Measures option to run the analysis. However, I believe you need to use the mean search time in each condition as your repeated measure variables, instead of the search times for each picture.

If you find a main effect of group, will that mean anything? I mean, was there a theory-based difference between the set of pictures you used for group 1 vs. group 2, 3, 4 ? If not, then you may also ignore the group variable and only run a repeated-measures ANOVA.

I hope this helps.

### Who is online

Users browsing this forum: No registered users and 1 guest