us housing affordability crisis housing crisis merged
Reading
This dataset covers 3,222 U.S. counties with 16 columns describing rental affordability — rents, incomes, renter shares, and burden percentages — keyed by FIPS and county name. Several numeric fields (annual_rent, median_gross_rent, median_household_income, rent_to_income_ratio) carry impossible negative sentinel values like -666666666 and -7999999992, which drag means deeply negative and produce skew around -17 to -57; these need cleaning before any analysis. The affordability_category field is also extremely imbalanced — 3,192 of 3,222 counties are labeled 'Affordable' (top_rate 0.99), so it offers little discriminatory signal as-is. The cleaner fields to start with are pct_rent_burdened_30plus (median 37.36%), pct_rent_burdened_50plus (median 17.62%), and pct_renter (median 26.07%), which look well-behaved and tell the real affordability story.
citing: annual_rent.stats.min · annual_rent.stats.skew · median_household_income.stats.min · median_household_income.stats.skew · rent_to_income_ratio.stats.skew · affordability_category.stats.top_rate · affordability_category.top_values · pct_rent_burdened_30plus.stats.median · pct_rent_burdened_50plus.stats.median · pct_renter.stats.median · row_count · column_count
Charts the summary said to look at first
Show data table
| bin | count |
|---|---|
| 0 – 1.624 | 9 |
| 1.624 – 3.248 | 5 |
| 3.248 – 4.872 | 3 |
| 4.872 – 6.496 | 5 |
| 6.496 – 8.12 | 9 |
| 8.12 – 9.744 | 13 |
| 9.744 – 11.37 | 11 |
| 11.37 – 12.99 | 16 |
| 12.99 – 14.62 | 26 |
| 14.62 – 16.24 | 19 |
| 16.24 – 17.86 | 35 |
| 17.86 – 19.49 | 43 |
| 19.49 – 21.11 | 52 |
| 21.11 – 22.74 | 52 |
| 22.74 – 24.36 | 73 |
| 24.36 – 25.98 | 99 |
| 25.98 – 27.61 | 109 |
| 27.61 – 29.23 | 116 |
| 29.23 – 30.86 | 132 |
| 30.86 – 32.48 | 159 |
| 32.48 – 34.1 | 189 |
| 34.1 – 35.73 | 209 |
| 35.73 – 37.35 | 227 |
| 37.35 – 38.98 | 239 |
| 38.98 – 40.6 | 205 |
| 40.6 – 42.22 | 209 |
| 42.22 – 43.85 | 210 |
| 43.85 – 45.47 | 190 |
| 45.47 – 47.1 | 131 |
| 47.1 – 48.72 | 114 |
| 48.72 – 50.34 | 118 |
| 50.34 – 51.97 | 69 |
| 51.97 – 53.59 | 51 |
| 53.59 – 55.22 | 34 |
| 55.22 – 56.84 | 24 |
| 56.84 – 58.46 | 6 |
| 58.46 – 60.09 | 3 |
| 60.09 – 61.71 | 2 |
| 61.71 – 63.34 | 3 |
| 63.34 – 64.96 | 3 |
Show data table
| bin | count |
|---|---|
| 0 – 1.624 | 42 |
| 1.624 – 3.248 | 27 |
| 3.248 – 4.872 | 34 |
| 4.872 – 6.496 | 63 |
| 6.496 – 8.12 | 102 |
| 8.12 – 9.744 | 148 |
| 9.744 – 11.37 | 163 |
| 11.37 – 12.99 | 214 |
| 12.99 – 14.62 | 242 |
| 14.62 – 16.24 | 310 |
| 16.24 – 17.86 | 315 |
| 17.86 – 19.49 | 332 |
| 19.49 – 21.11 | 335 |
| 21.11 – 22.74 | 264 |
| 22.74 – 24.36 | 219 |
| 24.36 – 25.98 | 150 |
| 25.98 – 27.61 | 99 |
| 27.61 – 29.23 | 64 |
| 29.23 – 30.86 | 39 |
| 30.86 – 32.48 | 20 |
| 32.48 – 34.1 | 21 |
| 34.1 – 35.73 | 9 |
| 35.73 – 37.35 | 2 |
| 37.35 – 38.98 | 3 |
| 38.98 – 40.6 | 1 |
| 40.6 – 42.22 | 1 |
| 42.22 – 43.85 | 1 |
| 43.85 – 45.47 | 0 |
| 45.47 – 47.1 | 1 |
| 47.1 – 48.72 | 0 |
| 48.72 – 50.34 | 0 |
| 50.34 – 51.97 | 0 |
| 51.97 – 53.59 | 0 |
| 53.59 – 55.22 | 0 |
| 55.22 – 56.84 | 0 |
| 56.84 – 58.46 | 0 |
| 58.46 – 60.09 | 0 |
| 60.09 – 61.71 | 0 |
| 61.71 – 63.34 | 0 |
| 63.34 – 64.96 | 1 |
Show data table
| bin | count |
|---|---|
| 3.01 – 5.435 | 1 |
| 5.435 – 7.859 | 3 |
| 7.859 – 10.28 | 9 |
| 10.28 – 12.71 | 26 |
| 12.71 – 15.13 | 63 |
| 15.13 – 17.56 | 156 |
| 17.56 – 19.98 | 316 |
| 19.98 – 22.41 | 371 |
| 22.41 – 24.83 | 450 |
| 24.83 – 27.26 | 419 |
| 27.26 – 29.68 | 357 |
| 29.68 – 32.11 | 301 |
| 32.11 – 34.53 | 203 |
| 34.53 – 36.96 | 169 |
| 36.96 – 39.38 | 115 |
| 39.38 – 41.81 | 75 |
| 41.81 – 44.23 | 56 |
| 44.23 – 46.66 | 45 |
| 46.66 – 49.08 | 25 |
| 49.08 – 51.5 | 15 |
| 51.5 – 53.93 | 11 |
| 53.93 – 56.35 | 10 |
| 56.35 – 58.78 | 8 |
| 58.78 – 61.2 | 4 |
| 61.2 – 63.63 | 4 |
| 63.63 – 66.05 | 1 |
| 66.05 – 68.48 | 1 |
| 68.48 – 70.9 | 3 |
| 70.9 – 73.33 | 1 |
| 73.33 – 75.75 | 1 |
| 75.75 – 78.18 | 0 |
| 78.18 – 80.6 | 1 |
| 80.6 – 83.03 | 0 |
| 83.03 – 85.45 | 1 |
| 85.45 – 87.88 | 0 |
| 87.88 – 90.3 | 0 |
| 90.3 – 92.73 | 0 |
| 92.73 – 95.15 | 0 |
| 95.15 – 97.58 | 0 |
| 97.58 – 100 | 1 |
Show data table
| value | count | share |
|---|---|---|
| Affordable | 3192 | 99.1% |
| Moderately Burdened | 29 | 0.9% |
| Extremely Burdened | 1 | 0.0% |
Show data table
| bin | count |
|---|---|
| -6.667e+08 – -6.5e+08 | 10 |
| -6.5e+08 – -6.333e+08 | 0 |
| -6.333e+08 – -6.167e+08 | 0 |
| -6.167e+08 – -6e+08 | 0 |
| -6e+08 – -5.833e+08 | 0 |
| -5.833e+08 – -5.667e+08 | 0 |
| -5.667e+08 – -5.5e+08 | 0 |
| -5.5e+08 – -5.333e+08 | 0 |
| -5.333e+08 – -5.167e+08 | 0 |
| -5.167e+08 – -5e+08 | 0 |
| -5e+08 – -4.833e+08 | 0 |
| -4.833e+08 – -4.667e+08 | 0 |
| -4.667e+08 – -4.5e+08 | 0 |
| -4.5e+08 – -4.333e+08 | 0 |
| -4.333e+08 – -4.167e+08 | 0 |
| -4.167e+08 – -4e+08 | 0 |
| -4e+08 – -3.833e+08 | 0 |
| -3.833e+08 – -3.667e+08 | 0 |
| -3.667e+08 – -3.5e+08 | 0 |
| -3.5e+08 – -3.333e+08 | 0 |
| -3.333e+08 – -3.167e+08 | 0 |
| -3.167e+08 – -3e+08 | 0 |
| -3e+08 – -2.833e+08 | 0 |
| -2.833e+08 – -2.667e+08 | 0 |
| -2.667e+08 – -2.5e+08 | 0 |
| -2.5e+08 – -2.333e+08 | 0 |
| -2.333e+08 – -2.167e+08 | 0 |
| -2.167e+08 – -2e+08 | 0 |
| -2e+08 – -1.833e+08 | 0 |
| -1.833e+08 – -1.667e+08 | 0 |
| -1.667e+08 – -1.5e+08 | 0 |
| -1.5e+08 – -1.333e+08 | 0 |
| -1.333e+08 – -1.167e+08 | 0 |
| -1.167e+08 – -1e+08 | 0 |
| -1e+08 – -8.333e+07 | 0 |
| -8.333e+07 – -6.666e+07 | 0 |
| -6.666e+07 – -5e+07 | 0 |
| -5e+07 – -3.333e+07 | 0 |
| -3.333e+07 – -1.666e+07 | 0 |
| -1.666e+07 – 2805 | 3212 |
Schema
16 columns| Alerts | ||||
|---|---|---|---|---|
| fips | numeric | 0.0% | 3,222 |
|
| county_name | text | 0.0% | 3,222 |
near_unique
|
| total_renters | numeric | 0.0% | 2,709 |
high_skew
outliers
|
| pct_rent_burdened_30plus | numeric | 0.0% | 2,146 |
|
| pct_rent_burdened_50plus | numeric | 0.0% | 1,769 |
|
| median_gross_rent | numeric | 0.0% | 984 |
high_skew
outliers
|
| median_household_income | numeric | 0.0% | 3,099 |
high_skew
outliers
|
| total_housing_units | numeric | 0.0% | 3,074 |
high_skew
outliers
|
| owner_occupied | numeric | 0.0% | 3,001 |
high_skew
outliers
|
| renter_occupied | numeric | 0.0% | 2,709 |
high_skew
outliers
|
| pct_renter | numeric | 0.0% | 1,925 |
|
| annual_rent | numeric | 0.0% | 984 |
high_skew
outliers
|
| rent_to_income_ratio | numeric | 0.0% | 1,278 |
high_skew
|
| affordability_category | categorical | 0.0% | 3 |
imbalance
|
| hours_at_min_wage_for_rent | numeric | 0.0% | 230 |
high_skew
outliers
|
| weeks_at_min_wage_for_rent | numeric | 0.0% | 72 |
high_skew
outliers
|
fips
numeric identifierThis is the U.S. county FIPS code, a 5-digit geographic identifier where the leading digits encode the state. Every one of the 3222 rows is unique with no nulls, and the range 1001 to 72153 spans Alabama through Puerto Rico, consistent with a full national county roster. Distribution stats (mean 31377.89, skew 0.157) are not meaningful here since the values are categorical codes, not quantities. Treatment: Treat as a categorical key; left-join on this code to bring in geographic attributes rather than using it as a numeric feature.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 3,222
- min
- 1,001
- max
- 72,153
- mean
- 3.138e+04
- median
- 30,022
- std
- 1.63e+04
- q1
- 1.903e+04
- q3
- 4.61e+04
- iqr
- 27,075
- skew
- 0.1574
- kurtosis
- -0.6314
- n_outliers
- 0
- outlier_rate
- 0
- zero_rate
- 0
county_name
text identifier near_uniqueThis column holds U.S. county identifiers, formatted as 'County, State' strings — 'county,' appears in 2999 of 3222 rows and the top state tokens (texas 256, virginia 189, georgia 159) match the U.S. counties-by-state distribution. 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 canonical name field. The 223 rows not containing 'county,' likely reflect Louisiana parishes, Alaska boroughs, or independent cities rather than data quality issues. Treatment: Use as a join key against county-level reference tables; do not feed raw into models.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 3,222
- len_min
- 16
- len_max
- 59
- len_mean
- 24.32
- len_median
- 24
- len_p95
- 31
- word_mean
- 3.248
- word_median
- 3
- n_empty
- 0
- n_duplicates
- 0
- duplicate_rate
- 0
- vocab_size
- 1,990
- readability_flesch_mean
- 10.28
- emoji_rate
- 0
- url_rate
- 0
- one_word_rate
- 0
- allcaps_rate
- 0
- boilerplate_rate
- 0
total_renters
numeric feature high_skew outliersThis 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 at some geography or entity level rather than a per-unit measure. The distribution is severely right-skewed (skew 15.82, kurtosis 398.15) with 449 outliers (14.0% outlier rate) and a mean (13,851) more than five times the median, indicating a few very large populations dominate. No nulls or zeros, and 2,709 unique values out of 3,222 rows suggest most records carry distinct counts. Treatment: log-transform before regression to tame the extreme right skew.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 2,709
- min
- 28
- max
- 1.811e+06
- mean
- 1.385e+04
- median
- 2580
- std
- 5.535e+04
- q1
- 1004
- q3
- 7396
- iqr
- 6,392
- skew
- 15.82
- kurtosis
- 398.2
- n_outliers
- 449
- outlier_rate
- 0.1394
- zero_rate
- 0
pct_rent_burdened_30plus
numeric featureThis appears to be the percentage of renter households spending 30%+ of income on rent, reported per geographic unit (likely county-level given n=3222). Values span 0 to 64.96 with a median of 37.36 and IQR of roughly 30.67–43.48, indicating rent burden is widespread rather than rare. The mild left skew (-0.57) and 58 outliers (1.8%) suggest a few areas with unusually low burden pull the tail down, while 0.25% report exactly zero. Treatment: Use as-is as a continuous feature; no transformation needed given near-symmetric distribution.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 2,146
- min
- 0
- max
- 64.96
- mean
- 36.44
- median
- 37.36
- std
- 10.01
- q1
- 30.67
- q3
- 43.48
- iqr
- 12.81
- skew
- -0.5673
- kurtosis
- 0.5032
- n_outliers
- 58
- outlier_rate
- 0.018
- zero_rate
- 0.002483
pct_rent_burdened_50plus
numeric featureThis is a county-level (or similar geographic) numeric feature giving the percentage of households spending 50%+ of income on rent — severely rent-burdened. Values span 0 to 64.96 with a near-symmetric distribution (skew 0.05) centered at a median of 17.62%, and the IQR of 8.56 around a mean of 17.35 indicates a tight, well-behaved spread. Only 47 outliers (1.46%) and a 0.93% zero rate, with no nulls across 3,222 rows, suggest clean and complete coverage. Treatment: Use as-is in modelling; standardize if combining with other scaled features.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 1,769
- min
- 0
- max
- 64.96
- mean
- 17.35
- median
- 17.62
- std
- 6.577
- q1
- 13.07
- q3
- 21.63
- iqr
- 8.557
- skew
- 0.05436
- kurtosis
- 0.9823
- n_outliers
- 47
- outlier_rate
- 0.01459
- zero_rate
- 0.009311
median_gross_rent
numeric feature high_skew outliersLikely the median gross rent (in dollars) for each row's geography, with a typical value near $817.5 and an interquartile range of $718–$978. The data is contaminated by sentinel values: the minimum is -666666666 and the mean of -2,068,220 with std ~3.7e7 is impossible for rent, producing extreme negative skew (-17.87) and kurtosis (317.2). 235 outliers (7.3%) are flagged, but the central distribution looks plausible once sentinels are removed. Treatment: Replace sentinel negatives (e.g., -666666666) with NaN before any modelling or aggregation.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 984
- min
- -6.667e+08
- max
- 2,805
- mean
- -2.068e+06
- median
- 817.5
- std
- 3.709e+07
- q1
- 718
- q3
- 978
- iqr
- 260
- skew
- -17.87
- kurtosis
- 317.2
- n_outliers
- 235
- outlier_rate
- 0.07294
- zero_rate
- 0
median_household_income
numeric feature high_skew outliersCounty-level median household income in dollars, with 3099 unique values across 3222 rows and no nulls. The distribution is contaminated by sentinel values: the minimum is -666666666 and the mean of -144603 is implausible against a median of 60458.5, driving extreme skew (-56.74) and kurtosis (3216.99). Once sentinels are removed, the IQR (51814.75 to 70376.25) looks like a typical US income distribution, but 188 outliers (5.83%) remain flagged. Treatment: Replace -666666666 sentinels with nulls, then consider log-transform or winsorization before modelling.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 3,099
- min
- -6.667e+08
- max
- 170,463
- mean
- -1.446e+05
- median
- 6.046e+04
- std
- 1.175e+07
- q1
- 5.181e+04
- q3
- 7.038e+04
- iqr
- 1.856e+04
- skew
- -56.74
- kurtosis
- 3217
- n_outliers
- 188
- outlier_rate
- 0.05835
- zero_rate
- 0
total_housing_units
numeric feature high_skew outliersCounts of total housing units across 3,222 rows with no nulls and 3,074 distinct values, consistent with one record per geographic area (e.g., county). The distribution is severely right-skewed (skew 12.05, kurtosis 240.5): the median is 10,021 but the mean is 39,403 and the max reaches 3,363,093, with 443 outliers (13.7%) above the IQR fence. Range spans 32 to 3.36M, indicating a mix of very small and very large jurisdictions. Treatment: log-transform before regression to tame the heavy right tail.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 3,074
- min
- 32
- max
- 3.363e+06
- mean
- 3.94e+04
- median
- 10,021
- std
- 1.201e+05
- q1
- 4211
- q3
- 25,939
- iqr
- 2.173e+04
- skew
- 12.05
- kurtosis
- 240.5
- n_outliers
- 443
- outlier_rate
- 0.1375
- zero_rate
- 0
owner_occupied
numeric feature high_skew outliersLikely a count of owner-occupied housing units per geographic area, given the integer-like range from 0 to 1,552,164 and median of 7,325.5. The distribution is severely right-skewed (skew 9.52, kurtosis 146.9) with 429 outliers (13.3%) and a mean (25,551.7) far above the median, indicating a few very large areas dominate. Near-unique values (3,001 of 3,222) and effectively no zeros (0.03%) suggest one row per area rather than a categorical flag. Treatment: log-transform before regression to tame the heavy right tail.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 3,001
- min
- 0
- max
- 1.552e+06
- mean
- 2.555e+04
- median
- 7326
- std
- 6.755e+04
- q1
- 3148
- q3
- 1.886e+04
- iqr
- 1.572e+04
- skew
- 9.516
- kurtosis
- 146.9
- n_outliers
- 429
- outlier_rate
- 0.1331
- zero_rate
- 0.0003104
renter_occupied
numeric feature high_skew outliersCounts of renter-occupied housing units per record, almost certainly aggregated by geography given the spread from 28 to 1,810,929. The distribution is extremely right-skewed (skew 15.82, kurtosis 398.15) with a median of 2,579.5 sitting far below the mean of 13,851, and 449 outliers (14% of rows) likely representing dense urban areas. Near-unique values (2,709 of 3,222) and zero null/zero rate suggest a clean per-area tally rather than a categorical feature. Treatment: log-transform before regression to tame the heavy right tail.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 2,709
- min
- 28
- max
- 1.811e+06
- mean
- 1.385e+04
- median
- 2580
- std
- 5.535e+04
- q1
- 1004
- q3
- 7396
- iqr
- 6,392
- skew
- 15.82
- kurtosis
- 398.2
- n_outliers
- 449
- outlier_rate
- 0.1394
- zero_rate
- 0
pct_renter
numeric featureNumeric share-of-renters feature spanning 3.01 to 100.0 with mean 27.35 and median 26.07, suggesting county- or tract-level renter percentages. The distribution is right-skewed (skew 1.32, kurtosis 4.41) with 88 high-side outliers (2.7%) pulling toward 100, while the IQR is a tight 10.02 around the mid-20s. No nulls or zeros, and 1925 unique values across 3222 rows indicate granular but not unique measurements. Treatment: Consider a log or Yeo-Johnson transform before regression to tame the right tail.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 1,925
- min
- 3.01
- max
- 100
- mean
- 27.35
- median
- 26.07
- std
- 8.564
- q1
- 21.64
- q3
- 31.66
- iqr
- 10.02
- skew
- 1.317
- kurtosis
- 4.412
- n_outliers
- 88
- outlier_rate
- 0.02731
- zero_rate
- 0
annual_rent
numeric feature high_skew outliersThis is an annual rent figure in currency units, with a typical tenant paying between 8616 and 11736 (median 9810). However, the minimum of -7999999992 drags the mean to -24818640 and produces extreme skew (-17.87) and kurtosis (317.20), indicating sentinel values or sign errors masquerading as rents. 235 outliers (7.3%) sit outside the IQR fence, so the tail is not a single rogue record. Treatment: Clip or null negative sentinels, then log-transform before modelling.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 984
- min
- -8e+09
- max
- 33,660
- mean
- -2.482e+07
- median
- 9,810
- std
- 4.451e+08
- q1
- 8,616
- q3
- 11,736
- iqr
- 3,120
- skew
- -17.87
- kurtosis
- 317.2
- n_outliers
- 235
- outlier_rate
- 0.07294
- zero_rate
- 0
rent_to_income_ratio
numeric feature high_skewThis is a numeric feature meant to capture rent as a proportion (or percentage) of income, with the bulk of values clustered tightly between q1 15.07 and q3 19.39 around a median of 17.05. However, the column is badly corrupted: the minimum is -24,357,569.09 driving a mean of -37,244.13 and a std of 752,361.70, with skew -22.74 and kurtosis 570.21. Negative ratios are nonsensical here and 114 outliers (3.54%) plus a max of 1200 suggest data entry or unit errors rather than genuine variation. Treatment: Clip or filter to plausible non-negative ratios and investigate the extreme negatives before any modelling.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 1,278
- min
- -2.436e+07
- max
- 1,200
- mean
- -3.724e+04
- median
- 17.05
- std
- 7.524e+05
- q1
- 15.07
- q3
- 19.39
- iqr
- 4.317
- skew
- -22.74
- kurtosis
- 570.2
- n_outliers
- 114
- outlier_rate
- 0.03538
- zero_rate
- 0
affordability_category
categorical label imbalanceA categorical affordability bucket with three levels: Affordable, Moderately Burdened, and Extremely Burdened. The distribution is extremely imbalanced — 'Affordable' covers 3192 of 3222 rows (top_rate 0.9907), leaving only 29 'Moderately Burdened' and a single 'Extremely Burdened' record. Entropy ratio of 0.049 confirms there is almost no information in this column as-is. Treatment: Collapse to a binary affordable-vs-burdened flag or drop; near-constant for modelling.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 3
- top_value
- Affordable
- top_rate
- 0.9907
- cardinality
- 3
- entropy
- 0.07815
- entropy_ratio
- 0.04931
hours_at_min_wage_for_rent
numeric feature high_skew outliersThis column appears to capture how many minimum-wage hours are needed to cover rent, with a typical value around 113 hours (IQR 99–135). However, the data is corrupted by extreme negative values: the minimum is -91,954,023 and the mean is -285,271, despite a max of 387, producing severe negative skew (-17.87) and kurtosis of 317. About 7.2% of rows (232) are flagged as outliers, suggesting sentinel codes or data-entry errors masquerading as numeric values. Treatment: Filter out negative sentinel values before use, then consider a log or robust transform.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 230
- min
- -9.195e+07
- max
- 387
- mean
- -2.853e+05
- median
- 113
- std
- 5.116e+06
- q1
- 99
- q3
- 135
- iqr
- 36
- skew
- -17.87
- kurtosis
- 317.2
- n_outliers
- 232
- outlier_rate
- 0.072
- zero_rate
- 0
weeks_at_min_wage_for_rent
numeric feature high_skew outliersThis column appears to capture how many weeks of full-time minimum-wage work are needed to cover rent, with a typical value around 2.8 weeks (IQR 2.5-3.4). However, the data is severely corrupted: the minimum is -2,298,850.6 and the mean is -7,131.79, while the max is only 9.7, producing extreme skew (-17.87) and kurtosis (317.20) with 232 outliers (7.2% of rows). Negative weeks are nonsensical for this metric, suggesting sentinel values or data-entry errors masquerading as numbers. Treatment: Filter out negative values as invalid before any modelling, then consider winsorizing the upper tail.
- n
- 3,222
- nulls
- 0 (0.0%)
- unique
- 72
- min
- -2.299e+06
- max
- 9.7
- mean
- -7132
- median
- 2.8
- std
- 1.279e+05
- q1
- 2.5
- q3
- 3.4
- iqr
- 0.9
- skew
- -17.87
- kurtosis
- 317.2
- n_outliers
- 232
- outlier_rate
- 0.072
- zero_rate
- 0