You can aggregate numerical values in one column for identical values in another column
For instance, for the cars dataset, you can first group for the year and the origin.
df = groupby(df, [:Year, :Origin])
Then, you can combine to get the average weight for each grouping
df = combine(df, :Weight_in_lbs => mean)
This is part of the so-called Split-Apply-Combine (SAC) pattern.