I'm sure this is an incredibly basic question to be asking on this forum, so I apologise in advance for that. I'm completely new to statistics. I have a dataset of interval data organised as follows: wine quality (on a scale from 0-100 by intervals of 10), sulphates (g(potassium sulphate)/dm3 - all of these fall between 0.330 and 2.00), and alcohol content (%, all of these fall between 8.4% and 14.9%). N = 1499. My aim is to predict whether sulphates and alcohol content are a good predictor of wine quality. To do this, I want to conduct a correlation analysis and a simple linear regression analysis. The variable tripping me up a lot is wine quality. My scatterplots all look like this: A scatterplot with units vertically "stacked" over the x-axis points And I don't know why. I've done a linear regression analysis but not with these kinds of variables.
I'm mostly just wondering why my scatterplots look like that and any tips on how to handle the wine quality variable. Thank you.
EDIT FOR CLARIFICATION: My basic questions are: (1) is simple linear regression the correct approach for this? i.e. Wine Quality = β0 + β1(Alcohol Content) + ε and Wine Quality = β0 + β1(Sulphates) + ε and (2) Is there any way to get my scatterplots to look normal?