You should encode watching frequency as a quantitative attribute. You are using a categorical one.
Then use frequency in the X axis, and repeat extraversion, conscientiousness and neuroticism in the y axis
For this idiom, you probably need to derive some better statistics. Maybe think of computing what is the average genre distribution that a person would watch, and then compute how is that distributed for people with high extraversion
Again, not very interesting...
Maybe if you could find the personality traits of directors it could be interesting to explore this.
However I think it is better for you to explore how movies are rated depending on different genres.
If you can find that for example Extroverts prefer comedies, then it could be interesting to see what ratings comedies normally get compared with the expected value