Still very very new to R! I've run a conjoint (choice) experiment that uses the function clogit. Output as follows: (this model has been obtained through stepwise methods)
clogout3a <- clogit(RES ~ ASC + Tasmania + Import + Price + strata(STR),data = dataset1)
> clogout3a
Call:
clogit(RES ~ ASC + Tasmania + Import + Price + strata(STR), data = dataset1)
coef exp(coef) se(coef) z p
ASC 2.096 8.135 0.1270 16.51 0.0e+00
Tasmania 0.457 1.580 0.1053 4.34 1.4e-05
Import -0.172 0.842 0.1330 -1.29 2.0e-01
Price -0.526 0.591 0.0505 -10.41 0.0e+00
Likelihood ratio test=512 on 4 df, p=0 n= 3402, number of events= 1134
> gofm(clogout3a)
Rho-squared = 0.2053607
Adjusted rho-squared = 0.20215
Akaike information criterion (AIC) = 1987.965
Bayesian information criterion (BIC) = 2008.099
Number of coefficients = 4
Log likelihood at start = -1245.826
Log likelihood at convergence = -989.9825
As an example, would I be corrected in saying that the odds of choosing a Tasmania product would be 1.58 times higher than the base (in this case a generic Australian product)? Or is it correct to say that a person is 1.58 times more likely to choose a product if it is a Tasmania product? (Or are this statements saying exactly the same thing??).
I've tried to find examples but have had no luck. Thanks in advance!
