A popular television show recently released a video preview for the upcoming season on its website. As fans of the show discover the video, the number of views of the preview video has grown each day during the last 2 weeks. The number of days since the release of the video and the natural log of the number of video views is shown in the scatterplot. A graph titled l n (views) versus days has days on the x-axis, and l n (views) on the y-axis. A line of best fit has positive slope.

A graph titled residuals versus days has days on the x-axis, and residual on the y-axis. Points are spread throughout the graph.

Based on the scatterplot and residual plot, what type of model would be appropriate for summarizing the relationship between days and number of views?

An exponential model is appropriate because the residual plot does not show a clear pattern.
A linear model is appropriate because the scatterplot shows a strong, positive linear relationship.
A power model is appropriate because the scatterplot of days and natural log of views is strong and linear.
An exponential model is appropriate because the relationship between days and the natural log of views is linear, and the residual plot does not show a clear pattern.