What test do I have to use?

#1
Hi everyone, thanks for reading and (maybe) helping me out. My promotor is on a sabbatical and I really have no clue what to do next.

In my research I ask my participants how often they watch some kind of 18+ movies, (1 = never -- 7 = almost everytime). In a follow up question I want to know how much these types of movies arouse them (1 = not exciting-- 5 = very much exciting). Now I want to know if there is some sort of association between how often people watch them and how much it arouses them.

Is there a way to analyse this. I thought maybe I could do correlations, but that won't tell me much. For example: hardcore has a Pearson correlation of .710 (sig <.001); but it doesn't say if they watch it often and enjoy it or they never watch it and don't enjoy it.

Thank you for your time
 

Karabiner

TS Contributor
#2
Now I want to know if there is some sort of association between how often people watch them and how much it arouses them.
That can be done using a Spearman rank correlation (more suitable than Pearson here, IMO,
but the interpretation is quite similar).

but it doesn't say if they watch it often and enjoy it or they never watch it and don't enjoy it.
So you want to know something different from "if there were some sort of association"?
Maybe you want to know how large the proportions are for different combinations?
How well that can be done depends on your sample size, amongst others.

With kind regards

Karabiner
 
Last edited:
#3
So you want to know something different from "if there were some sort of association"?
Maybe you want to know how large the proportions are for different combinations?
Yes, this exactly, thank you. I have a total N of 418, per category of movie the N is between 121 and 406.
I already know how many people watch every category and how arousing people find certain categories, but now I would like to know if people watch it more often, they will also find it more arousing or not.

I might have a hard time explaining what I mean because English is not my first language and I don't understand statistics
 

Karabiner

TS Contributor
#4
but now I would like to know if people watch it more often, they will also find it more arousing or not.
This is what the correlation tells you. If frequency of watching goes up, so does the arousal (and vice versa).
The association is quite strong (r=0.7).

Maybe you some visualization could be useful. You make a graph with 7 bars (one for each level of the "how often..."
variable), and in each bar you show the frequency or the % of respondents on this level who answered "not
existing" or "a little exiting" etc., similar to this or to this, respectively.

With kind regards

Karabiner