Determining which variable is the cause and which one is the effect

I'm working on my assignment and I'm a bit lost. So the assignment requires us to draw a scatter plot based on data that we collected online. The purpose of the assignment is to see if higher earnings are correlated with a higher percent of university graduates. We were instructed to plot the cause on the x-axis and the effect on the y-axis. I was planning on placing the percentage of the population that has a university degree on the y-axis and the earnings on the x-axis. However, looking at some scatter plots online they have it the other way around (isn't this a reverse cause). Are both methods correct or is mine wrong?

I also had a second question regarding this assignment. In a previous assignment I lost a lot of marks for making the scale of my scatter plot small. I'm unsure how the scale could have been small when the percentage that I was working with ranged from 0.6% to 1.4%. I was wondering, how should I go about making the scale for this scatter plot since my lowest data value is 0.2% and my highest data value is 0.4%?

Additional info: I used excel to make my scatter plot and the points did look squished together. I would love to do draw it by hand this time around so any help I can get on this is much appreciated.

• What support do you have for your theory that an individual's earnings (presumably post-graduation) influence whether they graduate? The second part of your question is off-topic here, but is answered by consulting the help for Excel's plots. – whuber Jan 6 at 18:44
• @ whuber, there's really no support for that claim. I was going to state that having a higher education does not cause one to have a higher earning. I was going to state that there could be a hidden variable (a common cause) for this relationship. As for my second question, thanks for letting me know. – Cece Jan 6 at 19:52

The purpose of the assignment is to see if higher earnings are correlated with a higher percent of university graduates.

Correlation does not imply causation, so both scatter plots should be valid.

We were instructed to plot the cause on the x-axis and the effect on the y-axis.

I'm not particularly pleased with this instruction. As I've noted above, correlation is not causation, and the "cause" could vary depending on the study. For example:

• Trying to classify a person's graduate status based upon their earnings
• Trying to predict earnings based on graduate status.

In the first case, it would be appropriate to place earnings on the X-axis, while in the second case, not so much.

Strictly speaking, I would say that it does not matter in this context which variable ends up on which axis. My intuition tells me that it might matter to your instructor, so I would send an email asking for clarification.

Like the comment said, the Excel portion is off topic, though I'd suggest using Googling Excel tutorials to help you out with that one.

Best of luck!

• Thank you so much. I was stressing over this and you've put me at ease. Although my teacher instructed us to plot the cause on the x-axis and the effect on the y-axis given what we have covered in class and what we talked about the whole semester I'm sure he doesn't want us to actually state that there is a cause-effect relationship but rather there could be a common cause. I don't know though. I could be completely wrong, but I will definitely ask him about that. Thanks again. As for me second question, I'll definitely do that. – Cece Jan 6 at 20:00