[R] compare grupos dichotomus dependent variable

[Jim Lemon]

Hi Aurora,
Perhaps what you are seeking is a test of proportions.

prop.test(c(sum(men),sum(women)),c(length(men),length(women)))

        2-sample test for equality of proportions with continuity correction

data:  c(sum(men), sum(women)) out of c(length(men), length(women))
X-squared = 1.0378e-30, df = 1, p-value = 1
alternative hypothesis: two.sided
95 percent confidence interval:
 -0.02903701  0.01046116
sample estimates:
     prop 1      prop 2
0.000000000 0.009287926

Warning message:
In prop.test(c(sum(men), sum(women)), c(length(men), length(women))) :
  Chi-squared approximation may be incorrect

If the question was "Have you had a baby in the past year?" the answer
is not too far off, but your sample is a bit small.

Jim


On Sat, Aug 8, 2015 at 3:09 AM, AURORA GONZALEZ VIDAL
<aurora.gonzalez2 at um.es> wrote:
> Hello everybody. I have a statistics question:
>
> let's say that I want to compaire answers between men and women to a yes/no
> question but I have so much more women than men, then, it looks like I
> cannot use chi squared test. Would it be correct to use U test (or ranked
> Wilcoxon test)?? What do you think?? The code is below, than you so much!!
>
> men<-rep( 0,12 )
> women <- c( 0,1,0,0,0,1,0,0,0,rep( 0,114 ),1,rep( 0,199 ) )
> wilcox.test( men, women )
> chisq.test( men, women )
>
>
> ------
> Aurora González Vidal
>
> Sección Apoyo Estadístico.
> Servicio de Apoyo a la Investigación (SAI).
> Vicerrectorado de Investigación.
> Universidad de Murcia
> Edif. SACE . Campus de Espinardo.
> 30100 Murcia
>
> @. aurora.gonzalez2 at um.es
> T. 868 88 7315
> F. 868 88 7302
> www.um.es/sai
> www.um.es/ae
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.