Hello everyone,

I have a hopefully solveable question, however, I would like to get your opinion on it, as I am not 100% sure. I have a dataset with product purchases. The purchases are split into 4 groups (Fat, Cereal, Dairy, Drink). I need to analyse the consumption (Volume x Amount of item purchased). My problem now is that the volume units vary. For instance Cereals are measured in grams, whereas Drinks are measured in mililiters etc. This means that I have to standardize it. I am now drawn between two options.

1.) Withdraw the mean of each observation and divide by standard diviation (Leads to an overall mean of 0 and sd. of 1)

2.) Or I simply divide consumption by mean (leads to a mean of 1)

What do you think is the right solution? When analyzing the distribution in a historgram, I could not see a significant difference...