saturn

This dataset covers 3,222 U.S. counties (one row per FIPS code) with population totals and uninsured counts and rates. Both total_pop and uninsured_pop are extremely right-skewed (skew 13.4 and 17.8) with hundreds of outliers, indicating a handful of very large counties dominate the raw counts — analysts should work in per-capita or log space. The uninsured_rate is the more comparable metric: median 0.12 with about 17.5% of counties reporting zero, and a long tail reaching 3.7 that warrants a data-quality check. The county_name field shows Texas, Virginia, and Georgia contributing the most counties, useful context for any state-level rollups.

This column is a 5-character FIPS code identifying U.S. counties, with every one of the 3222 rows holding a unique value (n_unique equals n) and zero nulls. Lengths are uniformly 5 (min, median, max all 5), values are single tokens (one_word_rate 1.0), and the leading samples like 01001, 01003, 01005 match Alabama county FIPS prefixes. It functions as a primary key rather than a feature.

This column holds U.S. county identifiers, likely formatted as 'County Name County, State' given that 'county,' appears in 2,999 of 3,222 rows and state names like Texas (256), Virginia (189), and Georgia (159) dominate the top tokens. Every one of the 3,222 rows is unique with zero nulls and zero duplicates, consistent with a canonical roster of U.S. counties. String lengths cluster tightly (min 16, median 24, max 59) and average 3.25 words, so formatting is highly regular.

Likely a county- or region-level total population count: 3222 rows with 3141 unique values, no nulls, integer-scale magnitudes from 47 up to 9,866,623. The distribution is extremely right-skewed (skew 13.38, kurtosis 298.69) with median 25,328 far below mean 102,232, and 14.06% of rows flagged as outliers. The std of 326,933 dwarfs the IQR of 54,579, consistent with a few massive metros pulling the tail.

Counts of uninsured people per record, likely aggregated to a geographic unit (3222 rows hints at US counties). The distribution is brutally right-skewed: median is 36 but the mean is 159.9 and the max hits 20915, with skew 17.8 and kurtosis 462.9. Roughly 17.2% of rows are zero and 11.4% flag as outliers, so a handful of large jurisdictions dominate the totals.

This looks like a per-record uninsured rate, ranging from 0.0 to 3.7 with a median of 0.12 and IQR of 0.21. The distribution is severely right-skewed (skew 4.10, kurtosis 27.70) with 230 outliers (7.14%) and 17.54% exact zeros, and the max of 3.7 is implausible if this is meant to be a proportion bounded at 1.