Do I need a chi-square test? Seems to fit the bill from what I've been reading.
I am wanting to perform a hypothesis test to see if the type of house people live in has any significant effect on the probability of the property having a bath installed. I have four housetypes: detached, semi-detached, flats and terraced with bath ownership rates of 97%, 91%, 88% and 94%, and sample sizes of 170, 147, 15 and 104 respectively.
My question is: what type of hypothesis test do I need to perform in order to determine whether or not the differences in bath ownership rates between the groups (house types) are significant? I'm used to dealing with numeric (interval) data for this sort of thing so would normally use t-tests or ANOVA but I don't think these are applicable to categorical data.
Actually isn't a chi-squared test an independance test for two categorical variables? And I have a set of categorical variables (house type) and a continuous variable (percentage of each house type that have a bath) so I take it chi-squared isn't suitable here?
Oh I don't know it's all too confusing...
Yes I thought of using a t-test but I have four groups (house types) not two so didn't think it was applicable. Also the dependant variable is a proportion (% of houses that have a bath) so this might make a difference, i.e. when I'm organising the data each house either has a bath or it doesn't (true or false, 1 or 0) so I'm not sure if this can be treated the same way as for example the number of people in each type of house (which would be continuous).
This video seems to imply I can use chi-squared:
YouTube - ‪Chapter 12: Chi-square: An Introduction [1of 2]‬‏