Five minutes with tip_usd in ride-hail-trips

Someone asked what is in tip_usd on ride-hail-trips, and the honest answer took one query.

8,000 rows, no nulls, 1,276 distinct values. Tip. Always 0 for cash trips in this city.

The five-number version: min 0.00, p10 0.00, median 1.96, p90 7.12, max 77.95. The mean is 3.13.

The p10 to p90 band — 0.00 to 7.12 — is where ordinary rows live, and it is the pair worth quoting when somebody asks what to expect. Min and max describe the two strangest rows in the table and nothing else; they are useful for spotting impossible values and misleading for everything else.

select
  count(*)                  as rows,
  count(tip_usd)            as present,
  count(*) - count(tip_usd) as nulls,
  count(distinct tip_usd)   as distinct_values
from ride_hail_trips;

The mean sits 60% above the median, which is the whole story: anything that reports the average of tip_usd is reporting a number most rows are below. Put the median next to it or drop the mean.

Where this bites: a filter like tip_usd > 3.13 reads as "above average" and selects a minority of rows — a smaller minority than the phrase suggests to whoever asked for it. If the request was "the typical ones", the threshold they meant was 1.96.

The grain is one row per completed trip, which is the context every one of those numbers depends on. None of them survive a change of grain, which is why "profile the column" and "profile the table" are the same job. Full schema, and the CSV, on the dataset page.