Number of Genres

When initially looking through the data set I noticed some pretty frivolous usage of genres. Seeing "The Big Lebowski" with 'Sports' among its genres made me consider the possibility that maybe tacking on more and more genres to a movie made it more searchable and, thus, more profitible. This graph shows that, up to a point this is very true. Though at 8 genres there is a tipping point where I guess it all gets too much.

An ANOVA resulting in a P value < 0.0001 showed that at least one of these groups is significantly different. A regression analysis might have also been appropriate, but I stuck with the ANOVA due to the tipping point.

When looking at ratings, however, I predicted movies to get poorer and poorer ratings, not only as the plots got too busy but as people were drawn into a movie which wasn't what they expected due to frivolous genre-ing. I was surprised to see that the more genres crammed into a story, the more highly people seem to rate it.

An ANOVA resulting in a P value < 0.0001 showed that at least one of these groups is significantly different.