I've collected feedback from different types of guests (families with kids, romantic couples, business partners) on about 50 restaurants. Each guest rated several aspects of his restaurant experience (service quality, comfort, food quality,...) on a likert scale. I've got about 30 ratings for each restaurant (each guest rated only one restaurant).
I would be curious to find out if there are some restaurants that are rated better or worse by specific user groups compared to others (for example, families with kids vs. business partners) but I'm not sure which statistical procedure to use... I hope someone can help me.
In particular, I would like to answer the following questions.
1) Do different types of users have different restaurant preferences?
Maybe I could do an ANOVA for each one of my 50 restaurants to see if the same restaurant is rated differently depending on user type, but how do I aggregate this data? How can I calculate the probability that a randomly selected restaurant will get different ratings depending on user type? I guess there's a more appropriate procedure...
2) Which restaurants are appreciated by a certain user type?
For example, how can I identify and group all restaurants that get better ratings from families than from other user types?
Many thanks in advance!
Advertise on Talk Stats