Selection of correct statistical test

#1
Hi all

My research involves testing email interventions to improve survey response rates (sent vs completed email survey). A sample of the output has been attached. The research question is to see if there is a significant difference between completion rates of the six interventions.

I am now confused about the statistical test to use - ANOVA vs chi-square vs Fisher's exact. With ANOVA, one of the variables needs to be continuous, but complete vs not complete is a categorical variable. In which case, I need to choose between chi-square or Fisher's exact. I could use some guidance on this, I am a bit lost!

Thanks in advance!
 

Attachments

Karabiner

TS Contributor
#2
From the description you gave, I'd suppose that this is a 6x2 crosstab which can be analysed using Chi².
Provided that you have number of expected (!) cases > 5 in each of the 12 cells of the table.

With kind regards

Karabiner
 
#3
Thank you Karabiner, that is helpful!

However, some interventions did turn up numbers lower than 5. Is Fisher's exact the method to be used instead?