It's appropiate because you get to see the change in your estimate from the actual value by seeing how far the point is from the line y = x.
If the estimates were correct then the scatter plot would look like a straight line with a slope of 1.
I used the Union Membership Percentage as the explanatory variable. I chose this because it seemed reasonable to expect that the percentage of union members would determine how many strikes and lockouts happened every decade.
I feel like the percentage of union members directly relates positively to the amount of stirkes and lockouts.
My second scatter plot shows how this information will look when we switch the variables.
As you can see, the relationship between the two variables with strikes and lockouts as the explanatory variable shows a positive direct relationship as well.
The bar graph helps us visualize the difference between the personal income per capita and the Governor's salary by state.
There seems to be a slight positive correlation between the two variables but it's not that great therefore I would not be confident presenting it as such.
There still seems to be a positive coreelation but once again it isn't that great of a correlation..
There seems to be only one piece of data (Connecticut) that is highest, but other than that, almost all the data is clustered.