2x3 contingency table

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

Gnomendanser
Posts: 1
Joined: Sat Dec 13, 2014 1:07 am

2x3 contingency table

Postby Gnomendanser » Sat Dec 13, 2014 1:31 am

Hi. I'm doing an assignment where i'm planning a hypothethical research project. In this project i'm supposed to find out how people who are diagnosed clinically depressed view the future. Are they optimistic or negative, etc. So in the table i have one variable with gender and one variable with the categories negative, a bit positive and positive. Does it make sense to use chi-square for measuring significance here? I know how to interpret it with a 2x2 table, but how can the source of difference be interpreted with more categories?

And are there any other ways i could do this? If i for example scored their outlook for the future on a scale, e.g 1-10? Would it then make sense to use other analysis to test significance?

Thanks for any answers, good people!
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: 2x3 contingency table

Postby GerineL » Mon Dec 15, 2014 9:29 am

This forum is not meant for homework, so you will probably not get a complete answer.
However, some pointers:

1. I think most people would measure "outlook on the future" with a continuous scale, that would mean you can use anova's or something like that. Think about whether you have a particular reason to use a 3-point scale.

2. to find out what type of analysis you should use, follow 2 steps:

step 1: figure out what type of variables you have (numeric/categorical; ordinal; interval; scale)
step 2: look up the appropriate test for those variables, for instance: http://www.ats.ucla.edu/stat/mult_pkg/whatstat/

Good luck!

Who is online

Users browsing this forum: No registered users and 2 guests

cron