Statistic analysis

I conducted a study to find out the users view of Apple Watch features. Population: 1. Native group: The data obtained from 23 participants through focus group text analysis ( we obtained 561 inputs (Negative + positive) from the 23 participants ( participant rate each feature of apple watch with negative or positive comment, each participant may rate more than one feature in each comment ) 2. Expert group : The data obtained from Reddit forum where people giving their opinion about apple watch features. We collected more than 1000 comments (each reviewer may have more than 2 or 3 input in same forum) but we find 624 comments are related to apple watch features rated by negative or positive) in that forum one expert participant may rate 2 or more features at the same time. 3. Need to check the normality of each group separately(Novice and Expert) 4. Then need to know is there any significance difference per each feature between two groups. 5. Need to know which group find that feature is more important even it is significant or not. 6. Also with data interpretation, I need from you to explain why do you use that method with evidence from any literature. *. I tried to use Goodness of fit for proportion to each group