Hi there. I'm a student doing a i'm doing a bivariate analysis of the relationship between budget (explanatory variable) and gross profit (response variable). The graph can be seen here. Image - TinyPic - Free Image Hosting, Photo Sharing & Video Hosting I'm just wondering, to improve my model do I remove the data points circled in black, or the data point circled in red, or both?. Also I picked a linear trendline. Is this appropriate? Thanks in advance. Also, is there a causal relationship between Budget of movies and gross profit of movies or a correlation? I think its corellation because there is no distinct causal relationship that exist as the occurrence of Budget does no cause the second (Gross profit). The first event is the cause and the second event is the effect. We can't say that as Budget increases, Gross profit increases. Although these two variables are positively correlated, I can't say that Budget causes Gross profit as there are other factors that could result in higher Gross profit for example lack of a excellent concept story of the film. ~not too sure if this is right :S. Help please Cheers