I would say there are two different approaches you can take in this type of problems. The first approach is to actually look at the scatterplot between explanatory and response variable. If the scatterplot appears to suggest a linear correlation between explanatory and response variable, a line would be an appropriate model.
The second approach is to look at the residual plot. If there is no notable U-pattern in the residual plot and the points appear to be random, it means that a line is an appropriate model to use for these data.
You can use either one of these approaches, or if you are able to argue using both approaches, that is even better.
Hi,
Thank you for your question.
I would say there are two different approaches you can take in this type of problems. The first approach is to actually look at the scatterplot between explanatory and response variable. If the scatterplot appears to suggest a linear correlation between explanatory and response variable, a line would be an appropriate model.
The second approach is to look at the residual plot. If there is no notable U-pattern in the residual plot and the points appear to be random, it means that a line is an appropriate model to use for these data.
You can use either one of these approaches, or if you are able to argue using both approaches, that is even better.
Does this answer your question?