Build a two-way summary from CSV categories
A two-way summary answers how one category breaks down across another without changing the source records.
Choose two meaningful dimensions
Select one field for output rows and a different field for output columns. Exact labels remain separate, so North and north are different categories. Blank or whitespace-only labels are grouped as a blank category.
Match the metric to the question
Record count measures how many records fall into each pair. Sum and average read one selected measure column as finite decimal text. Blank measures are skipped; a nonblank label such as unknown stops the numeric calculation so the result does not hide malformed values.
Read empty combinations carefully
An unobserved pair has a count of zero and a sum of zero. Its average is blank because there is no value to average. A blank column label is shown as (blank) in the header; a blank row label remains an empty cell.
Check the output size and source totals
The number of distinct column labels determines the output width, and each row label creates an output record. Compare a few category pairs and totals with the source before using the summary. Very many distinct labels may exceed the site output limit.