Scattered plot- Alteryx advanced question help

[cerchiara]

I had a question on the Alteryx advanced exam where I needed to look at a graph and select what options were selected in the scatterplot tools configuration. However I'm having difficulty understanding what each tool is doing. Can someone help me understand these tools better or point me to the right direction?

Smooth line- This one I'm not to sure about. Attached I created a Alteryx scattered plot graph, and I can see a dotted line, is that all it is?

Marginal boxpoints- This one is a easy one as you can see the box points outside of the graph

Log x axis- Not totally sure how you can see this by just looking at the graph

Log y axis- Not totally sure how you can see this by just looking at the graph

Below are the definitions but something visual showing the differences would help me understand this better. I tried this by creating my own scatterplot Alteryx graph in the attached but the only one that I can be truly sure about is the Marginal boxpoints.

Log X axis: If selected, a natural log transformation is applied to the X values. Doing this is often useful for exploring certain types of non-linear relationships. Log Y axis: If selected, a natural log transformation is applied to the Y values

Smooth line: Displays a non-linear line between the X and Y fields that is created using a loess (non-parametric local regression) model.* Included by default

[OllieClarke · Accepted solution]

Hi @cerchiara 

If you look at this graph from the wikipedia page on logarithmic scales:

You can see that each increment on the Y axis is 10x bigger than the one before. So equal distances on a log scale are actually displaying 10x the distance in value. They can be really useful when looking at exponential growth (if you remember the Covid growth charts from a few years ago).

Another clue is that with log gridlines, they aren't always equally spaced, as you can see in the below example:

But generally look at the grid lines and values to identify if it's logged or not

The smooth line is not straight, unlike the least-squares (regression) line, so that should help visually identify it

Hope that helps,

Ollie

[cerchiara]

Thanks @OllieClarke , this makes sense. But it didn't seem clear on the exam. If you take the sales vs variable 1 graph and enabled and disabled the log x and log y options are the values of the x and y axis supposed to be changing? Or is it just the scattered points that are? On the alteryx file I attached if I enable and disable log x for example it just looks like the scattered points move from off the line to the line. 

[OllieClarke]

Hey @cerchiara 

So the data and values don’t change. It’s just the way they look that does. 

if I’ve got x values of 1,2,3 

and y values of 1,10,100

if i graph them normally they look very spread out, with a connecting line that curves sharply upward. 

if I log the y axis, then they are equally spaced with a straight connecting line. 

the values are still the same though

(please excuse my terrible drawing)

 We’re just changing what a visually equal distance means

[cerchiara]

My drawing skills would look about the same ha! This helps a lot though thanks! I assume for the Log X then its similar but opposite then?  

[OllieClarke]

@cerchiara 

Yup, exactly the same, just on the bottom instead of the side