Two large beach holiday destinations near Bangkok, Thailand are Pattaya and Hua Hin. In a random sample of 100 individuals in Bangkok who have been on a weekend away to nearby beach destinations 6 preferred Pattaya and 3 preferred Hua Hin. The other 91 did not have a preference. These individuals were then exposed to certain tourism promotional material and after such exposure 12 preferred Pattaya and 1 preferred Hua Hin. The remaining 87 had no preference. I would like to set up a hypothesis test to determine if the promotional material had an effect on preference. Your assistance or a referral to a URL for a tutorial on this kind of hypothesis testing would be greatly appreciated.
You can apply a chi-squared Test of independence test, but you will face an issue with the fact that only 1 respondent has chosen Hua Hin post receiving a stimulus.
Your alternative would be to run a Fisher's exact test to accommodate less than a frequency of 5 in each cell.
Note: Since your samples are paired, you would want to use a McNemar test instead.