How to make a chart with two Values and inverse proportionality?

My first thought for illustrating trends/relationships between two numeric values is to plot the data in an XY scatter chart. Does that show what you want?

I am not sure that I follow you. I assume that you are NOT referring to the X-axis and the Y-axis when you talk about x and y. If you are referring to the relationship between the two data sets (blue and brown), then surely the relationship is not proportionate, but inversely proportionate??? Couldn't you plot one data set against the other???

Without having any data to play with... that's all I will do. Unless you want to attach an Excel sheet (rather than a picture of one).

Unfortunately the attachment icon doesn't work at the moment. So, to attach an Excel file you have to do the following: Just before posting, scroll down to Go Advanced and then scroll down to Manage Attachments. Now follow the instructions at the top of that screen.

I am also comparing this graph to other graphs wich are directly proportionate, so it might confuse the reader if I depict them like this.

Do you have examples of these other graphs?

Have you definitely established that there is a correlation between x and y in the first example? The main difference that I see between the two data sets is that the second one seems to show a clear relationship/correlation between x and y, where the first example seems to show that there may not be a correlation between x and y. A quick CORREL() function on the two data sets shows an r of -0.63 (r^2=0.4) for the first set and +0.85 (r^2=0.73) for the second set -- suggesting that the correlation in the first set is quite a bit weaker than in the second set -- which is what the charts are also showing.

Is your goal here to manipulate the charts so that they (in particular the first one) show a stronger correlation than may really exist?

My inclination at this point is to suggest that the charts are doing just fine. They are showing those x-y variables that show stronger and weaker correlations in a fairly straightforward way.

I know that, sometimes in data analysis, we are trying to force the data to show us something that it doesn't really show. For example, here is a list of "spurious correlations" that are obviously meaningless: http://www.tylervigen.com/spurious-correlations

One thing you might consider to perhaps help highlight the negative correlation is to select "values in reverse order" (Format axis dialog) for one of the axes in those charts that show a negative correlation. As you have them, negative correlation shows up when one variable increases as the other decreases. If you format one of the axes to be high to low (instead of low to high), then the inverse correlation will appear more "parallel" -- like those that show positive correlation. It might be more a question of which creates more confusion -- crossing/mirror shaped lines with normal axes or parallel lines but opposite direction axes.