It sounds like multiple regression. I'd run two analyses, one for each IV.
Predictors are sociosexuality + your three constructs.
Are you also thinking about interactions?
J
On 30 July 2013 03:23, LIDBORG Linda <[log in to unmask]> wrote:
> Hi all,
>
> I'm somewhat confused about which test to use/how to go about the analysis
> of my data, and would be really grateful for some advice.
>
> I've carried out an eye tracking study investigating the eye gaze movements
> of males assessing female attractiveness from photographs. I'm particularly
> interested in how these are affected by males sociosexuality (their tendency
> to engage in and interest in casual, uncommitted sex). My hypothesis is that
> males who score highly on a sociosexuality construct will look relatively
> more at females bodies compared to males low in sociosexuality.
>
> To extend my study, I'm also investigating whether self-perceived mate value
> and narcissistic tendencies have an influence, so I've administered three
> constructs and I'm using each participants mean score on each construct. The
> thre constructs show moderate, positive correlations with each other.
>
> As for the eye tracking data, after inspecting data and carrying out
> preliminary analyses, I've chosen 2 DVs. The first one is time spent
> fixating on the fave versus the body, expressed as a face to body bias,
> where a negative value denotes more time fixating on the body and a positive
> value denotes more time fixating on the face. The strength of that
> preference is of course also expressed by the size of the value.
>
> The second DV is the time it took the before the participant looked at the
> face versus the body, also expressed as a face to body bias, negative value
> = looked at the body first, positive value = looked at the face first.
>
> Now, I've looked at the correlations, and all three constructs show
> moderate, negative correlations (sig) with DV 1, and one of the constructs
> show a moderate, negative correlation with DV 2 (the other two aren't sig),
> but I don't think this is sufficient? Should I also be using a multiple
> linear regression model?
> Any help would be very much appreciated!
>
> Thank you,
> Linda Lidborg
> MSc Psychologocial Research
> University of Edinburgh
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
|