saturn

This dataset contains 3,222 rows describing U.S. county-level poverty, with three columns: a FIPS code, a county name, and a poverty rate. Each row is a unique county (3,222 unique FIPS codes and county names), so the analytical signal lives in the poverty_rate column. Poverty rates range from 1.6% to 66.32% with a mean of 15.1% and median of 13.55%, and the distribution is right-skewed (skew ≈ 2.10) with 137 outliers on the high end. The county_name field also reveals geographic concentration, with Texas (256), Virginia (189), and Georgia (159) contributing the most counties. Start by examining the shape of poverty_rate and which states the high-poverty outliers cluster in.

This column holds 5-character FIPS codes uniquely identifying each of the 3222 rows (n_unique equals n, null_rate 0). Every value is exactly 5 characters, one word, all-caps/numeric, with zero duplicates or empties. Sample values like 01001, 01003, 01005 match the standard US county FIPS encoding (state+county).

This is a county identifier string, likely formatted as " County, " — "county," appears in 2999 of 3222 rows and Texas (256), Virginia (189), and Georgia (159) lead the state mentions. Every one of the 3222 values is unique with zero nulls or duplicates, and lengths cluster tightly (min 16, median 24, max 59), consistent with a clean US county roster. The 223 rows lacking the "county," token are worth checking — likely parishes, boroughs, or independent cities.

This is a county- or tract-level poverty rate expressed as a percentage, ranging from 1.6 to 66.32 with a median of 13.55 and IQR of 7.75. The distribution is right-skewed (skew 2.10, kurtosis 6.89) with 137 high outliers (4.25%) reflecting pockets of severe poverty well above the typical 10–18% band. No nulls or zeros, and 1719 unique values across 3222 rows suggest one record per geographic unit.