Friday, November 13, 2015

Do Stephen King's Better Books Make Better Films?

My last post (and data) got a bit popular on reddit. Some people noticed that Stephen King films ratings and the books ratings seemed highly correlated.

I think Max is right on this. So I made a graph showing how movie and book ratings are correlated

Now the actual correlation figure is


 cor(good.doubled, imdb, use="complete")
[1] [1] 0.8766 so it looks like highly rated King books make highly rated films

It could be that a good film makes people read and rate highly a book. But my basic conclusion is Stephen King's highly rated books make higher rated films















Appendix: Code for the Graph


mydata = read.csv("King.csv")
attach(mydata)
cor(good.doubled, imdb, use="complete")
p1 <- ggplot(mydata, aes(x=good.doubled, y=imdb)) +
    geom_point(shape=1) +    # Use hollow circles
    geom_smooth(method=lm,   # Add linear regression line
                se=FALSE)    # Don't add shaded confidence region
p1 <- p1 + ylab("IMDB Ratings")  
p1 <- p1 + xlab("GoodReads Ratings *2")  
p1 <- p1 + ggtitle("Stephen King: Books vs. Movies Correlation")
p1 <- p1 + geom_text(aes(label=ifelse(good.doubled>0,as.character(Title),'')),hjust=0,just=0,size=2, position = "jitter")
p1
ggsave("correlate.png")





Other Correlations I changed the publication data column into a column of Years since 1974.


 cor(good.doubled, Years, use="complete")
[1] -0.1879052 Not a strong correlation about Kings adapted books get better or worse since he started
> cor(good.doubled, Pages, use="complete")
[1] -0.02715659 no relationship between the length of a book and it being rated highly or lowly.
> cor(Years, Pages, use="complete")
[1] 0.482048 King's adapted books may be getting a bit longer over time

No comments: