saturn

This dataset contains 3,222 rows of U.S. county-level rent burden statistics, with each row identified by a county name and FIPS code and described by total renters and the share of renters paying 30%+ or 50%+ of income on rent. Total renters is extremely skewed (skew 15.8, max 1,810,929 vs. median 2,579.5), so a handful of large urban counties dominate the distribution and warrant separate treatment. Rent-burden percentages are more well-behaved: about 36.4% of renters per county are cost-burdened at the 30%+ threshold and 17.4% at the 50%+ threshold on average, both fairly symmetric. The most useful first look is comparing the two rent-burden distributions and isolating the outlier counties on total_renters.

This is the FIPS county/state code, used as a unique geographic identifier — every one of the 3,222 rows has a distinct value with no nulls. The range from 1001 to 72153 and the low skew (0.157) reflect the standard FIPS numbering across U.S. states and territories rather than a meaningful numeric distribution. Treating these as numbers (mean 31,377, std 16,299) is misleading; they are categorical codes.

This column holds fully-qualified US county names (e.g., 'X County, State'), with 3222 rows all unique and zero nulls. The token 'county,' appears 2999 times, suggesting ~223 entries don't follow that exact pattern — likely Louisiana parishes, Alaska boroughs, or independent cities worth checking. State frequencies match expectations, with Texas (256) leading.

This is a numeric count of renters per record, ranging from 28 to 1,810,929 with a median of 2,579.5 — likely an aggregate count at some geographic or entity level. The distribution is extremely right-skewed (skew 15.82, kurtosis 398.15) with the mean (13,851) over five times the median, and 449 outliers (13.9%) inflate the std to 55,351. No nulls or zeros, and 2,709 unique values across 3,222 rows suggest minor repetition but largely distinct totals.

This column captures the percentage of households spending 30%+ of income on rent, reported per row across 3,222 records with no nulls. Values span 0 to 64.96 with a median of 37.36 and IQR of 30.67-43.48, mildly left-skewed (-0.57) and tightly clustered (std 10.0). About 0.25% are exact zeros and 58 rows (1.8%) flag as outliers, but the distribution is otherwise well-behaved and ready to use as-is.

Likely the percentage of households spending 50%+ of income on rent at some geographic unit (e.g., county). Values span 0 to 64.96 with mean 17.35 and median 17.62, a near-symmetric distribution (skew 0.05) and modest tails (kurtosis 0.98). Only 0.93% are zero and 1.46% flagged as outliers, so the signal is clean and ready to use.