saturn

This dataset contains 3,222 rows of US county-level poverty statistics, with each row identified by a FIPS code, county name, and state abbreviation, plus three poverty rate measures and a population total. The poverty measures are all right-skewed: pct_poverty ranges from 1.6% to 66.32% with a median of 13.55%, while pct_deep_poverty has a median of 5.82% but reaches as high as 34.7%. The total population column is extremely skewed (skew of 13.4, kurtosis ~297) with a median of 25,174 but a max near 9.8 million, so any aggregate analysis should account for this. Texas (254 counties), Georgia (159), and Virginia (133) dominate the state distribution, which matters for any state-level rollups.

This column is the US county FIPS code: every one of the 3222 rows is unique, null-free, and the value range (1001 to 72153) matches the standard 5-digit state+county encoding. Treating it as numeric is misleading despite the clean distribution (skew 0.16, no outliers) — the digits are categorical identifiers, not measurements.

This column holds US county-level place names — virtually every value ends in 'County' (2999 occurrences), with smaller groups of Louisiana 'parish' (64) and Puerto Rican 'municipio' (78) entries. Despite 3222 rows, only 1960 are unique and 39.2% are duplicates, because common names like Washington County (30), Jefferson County (25) and Franklin County (24) recur across states. Values are short and uniform (mean 14.2 chars, ~2 words), so the name alone does not uniquely identify a county.

This is a US state code field with 52 distinct values, consistent with the 50 states plus DC and likely one territory. Distribution is broad and near-uniform (entropy ratio 0.93), with TX leading at just 7.88% (254 of 3222 rows), followed by GA, VA, KY, and MO. No nulls, and the row count suggests multiple records per state rather than one-per-state.

A heavily right-skewed numeric measure (skew 13.36, kurtosis 297.59) ranging from 47 to 9,782,602 with a median of 25,174 but a mean of 101,340 — the upper tail dwarfs the center. Roughly 13.9% of rows (449) flag as outliers, and the standard deviation (324,628) is over three times the mean, signalling a few very large values dominate. With 3,173 unique values across 3,222 rows and no nulls or zeros, this looks like a per-record aggregate total rather than a category or flag.

This is a numeric feature representing the percent of population in deep poverty, likely at a county or similar geographic unit (n=3222 with 1131 unique values). The distribution is right-skewed (skew 2.67, kurtosis 10.4) with a median of 5.82 but a max of 34.7, and 176 outliers (5.46%) sit in the upper tail. Min is 0.0 but the zero rate is just 0.09%, so the floor is rarely hit.

Likely a county- or tract-level poverty rate expressed as a percentage, ranging from 1.6 to 66.32 with a median of 13.55. The distribution is heavily right-skewed (skew 2.10, kurtosis 6.89) with 137 high-end outliers (~4.3%) pulling the mean (15.10) above the median. No nulls or zeros across 3,222 rows.

This column reports a percentage of population near the poverty line, ranging from 0.58 to 49.14 with a mean of 9.81 and median of 9.38. The distribution is right-skewed (skew 1.19, kurtosis 5.73) with 82 outliers (2.55%) on the high tail, but no nulls or zeros. The IQR is tight at 4.43, so most observations cluster between 7.33 and 11.76 with a long upper tail.