Having hard time coming up with an null hypothesis for one way Anova

My independent variable is assessment of whether respondents fear that crime is the worst problem in their city, and this assessment has 3 catagories (yes, uncertain, and no). I am looking at a data set in SPSS which lists all USA cities and many variables. My dependent variable are either number of single parent families in USA, number of black people in USA, number of people living below the poverty line. I can't say "the assessment affects number of people living below the poverty line". I thought about this for a long time, but can';t come up with a meaningful null hypothesis. Can you guys help? Thanks very much in advance.


Fortran must die
I dont see how you can possible have the IV you listed (fear of crime) and have dependent variables such as the number of blacks or people living under poverty. By that logic how afraid you were (or weren't) would determine your race or your income and neither seems logical (or even physically possible). I think the logic would go for example RACE -> Fear of Crime not Fear of Crime -> Race


Omega Contributor
What is your purpose? Are you just wanting to play around with anova? If so pick any categorical and continuous variables Evan f they don't make since. However interpreting the results beyond that may be impossible or no feasible.