Which test to conduct?

[Jennica]

Hi everyone, new here so please bare with me

Okay, so I'm currently conducting some research. My non-directional hypothesis is 'Does the media effect Christian women and their perception of body image?'

My two variables are age and ethnicity.

I'm absolutely clueless on which test is best used to analyse. I initially thought paired-sample t-test, but would this work considering all they have done is answer a questionnaire on one separate occasion? And my participants are only Christian women.

One more thing, the two questionnaires I used have sub scales, if I was to correlate the means of each sub scale from one questionnaire to another sub scale in the other questionnaire, would they really provide any insightful results?

I really appreciate all your help! Thank you in advance! :)

[Rich Ulrich]

This is not an SPSS question.  It appears to be a question about
trying to devise a question for a basic research class.  The term
that comes to my mind is "operationalize".   If you are going to
study something, you need to measure it.

(In a sample of Christian women,) Do media affect perception
of body image?  Well, probably so.

What are "media" that impinge on people differentially, that you
measure in some way?  You won't find people who have never
been exposed to TV, radio, movies, books ...  so what is being
measured?  - You probably want a clearer, simpler statement of
hypothesis, focusing on what you can actually measure.

Affecting "body image" is one thing, which might be presumed
to be one's own body image.  What does it mean to "affect
perception of body image"?  I suppose, if "body image" could
mean that "I think I am fat", then "perception of body image" might
mean, "I think I (or you) am fat, and that is a bad thing."
Is this a matter of increased/decreased stereotyping in the viewer?
Is it "tolerance" that is encouraged or discouraged by increased
exposure to some particular medium?  - Again, you probably want
a clearer, simpler statement of something you can actually
measure.

After you establish that there is a connection between the two, then
you can start to see whether that is influenced by your two variables
of age and ethnicity.

--
Rich Ulrich

> Date: Tue, 2 Apr 2013 06:11:03 -0700

> From: [hidden email]
> Subject: Which test to conduct?
> To: [hidden email]
>
> Hi everyone, new here so please bare with me
>
> Okay, so I'm currently conducting some research. My non-directional
> hypothesis is 'Does the media effect Christian women and their perception of
> body image?'
>
> My two variables are age and ethnicity.
>
> I'm absolutely clueless on which test is best used to analyse. I initially
> thought paired-sample t-test, but would this work considering all they have
> done is answer a questionnaire on one separate occasion? And my participants
> are only Christian women.
>
> One more thing, the two questionnaires I used have sub scales, if I was to
> correlate the means of each sub scale from one questionnaire to another sub
> scale in the other questionnaire, would they really provide any insightful
> results?
>
> I really appreciate all your help! Thank you in advance! :)
>
>
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Which-test-to-conduct-tp5719230.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