saturn·

← all readings

housing units

source /home/coolhand/html/datavis/data_trove/cache/housing_units.parquet 3,222 rows 6 columns profiled 2026-05-01 raw JSON static .html .ipynb Report Notebook

Reading

dataset summary · high confidence anthropic:claude-opus-4-7

This dataset covers 3,222 U.S. counties with housing-unit counts (owner-occupied, renter-occupied, total) plus a FIPS code, county name, and the percent of renters. The three count columns are extremely right-skewed (skew between 9.5 and 15.8, kurtosis above 140) with 13–14% of rows flagged as outliers — a handful of huge urban counties (max total_housing_units of about 3.36M vs a median of roughly 10,021) dominate the distribution. The pct_renter field is far better behaved, centered near 26% with a much tighter spread, making it the most useful comparable metric across counties. Start by inspecting the long tail of total_housing_units, then use pct_renter to compare counties on a normalized basis.

citing: owner_occupied.stats · renter_occupied.stats · total_housing_units.stats · pct_renter.stats · fips.stats · county_name.top_words · row_count

total_housing_units · Heavily right-skewed: most counties cluster near the median of ~10K units while a few exceed 3M.
Show data table
bincount
32 – 8.411e+042907
8.411e+04 – 1.682e+05153
1.682e+05 – 2.523e+0562
2.523e+05 – 3.363e+0538
3.363e+05 – 4.204e+0522
4.204e+05 – 5.045e+056
5.045e+05 – 5.886e+0511
5.886e+05 – 6.726e+055
6.726e+05 – 7.567e+055
7.567e+05 – 8.408e+053
8.408e+05 – 9.249e+051
9.249e+05 – 1.009e+063
1.009e+06 – 1.093e+061
1.093e+06 – 1.177e+061
1.177e+06 – 1.261e+060
1.261e+06 – 1.345e+060
1.345e+06 – 1.429e+060
1.429e+06 – 1.513e+060
1.513e+06 – 1.597e+060
1.597e+06 – 1.682e+061
1.682e+06 – 1.766e+061
1.766e+06 – 1.85e+060
1.85e+06 – 1.934e+060
1.934e+06 – 2.018e+060
2.018e+06 – 2.102e+061
2.102e+06 – 2.186e+060
2.186e+06 – 2.27e+060
2.27e+06 – 2.354e+060
2.354e+06 – 2.438e+060
2.438e+06 – 2.522e+060
2.522e+06 – 2.606e+060
2.606e+06 – 2.69e+060
2.69e+06 – 2.775e+060
2.775e+06 – 2.859e+060
2.859e+06 – 2.943e+060
2.943e+06 – 3.027e+060
3.027e+06 – 3.111e+060
3.111e+06 – 3.195e+060
3.195e+06 – 3.279e+060
3.279e+06 – 3.363e+061
pct_renter · A relatively well-behaved distribution centered around 26%, useful for cross-county comparison.
Show data table
bincount
3.01 – 5.4351
5.435 – 7.8593
7.859 – 10.289
10.28 – 12.7126
12.71 – 15.1363
15.13 – 17.56156
17.56 – 19.98316
19.98 – 22.41371
22.41 – 24.83450
24.83 – 27.26419
27.26 – 29.68357
29.68 – 32.11301
32.11 – 34.53203
34.53 – 36.96169
36.96 – 39.38115
39.38 – 41.8175
41.81 – 44.2356
44.23 – 46.6645
46.66 – 49.0825
49.08 – 51.515
51.5 – 53.9311
53.93 – 56.3510
56.35 – 58.788
58.78 – 61.24
61.2 – 63.634
63.63 – 66.051
66.05 – 68.481
68.48 – 70.93
70.9 – 73.331
73.33 – 75.751
75.75 – 78.180
78.18 – 80.61
80.6 – 83.030
83.03 – 85.451
85.45 – 87.880
87.88 – 90.30
90.3 – 92.730
92.73 – 95.150
95.15 – 97.580
97.58 – 1001
owner_occupied · Long-tailed counts with ~429 outlier counties pulling the mean far above the median.
Show data table
bincount
0 – 3.88e+042761
3.88e+04 – 7.761e+04225
7.761e+04 – 1.164e+0578
1.164e+05 – 1.552e+0552
1.552e+05 – 1.94e+0536
1.94e+05 – 2.328e+0520
2.328e+05 – 2.716e+0510
2.716e+05 – 3.104e+0510
3.104e+05 – 3.492e+056
3.492e+05 – 3.88e+056
3.88e+05 – 4.268e+053
4.268e+05 – 4.656e+053
4.656e+05 – 5.045e+054
5.045e+05 – 5.433e+052
5.433e+05 – 5.821e+050
5.821e+05 – 6.209e+051
6.209e+05 – 6.597e+051
6.597e+05 – 6.985e+050
6.985e+05 – 7.373e+050
7.373e+05 – 7.761e+050
7.761e+05 – 8.149e+050
8.149e+05 – 8.537e+050
8.537e+05 – 8.925e+050
8.925e+05 – 9.313e+051
9.313e+05 – 9.701e+050
9.701e+05 – 1.009e+060
1.009e+06 – 1.048e+060
1.048e+06 – 1.087e+061
1.087e+06 – 1.125e+060
1.125e+06 – 1.164e+060
1.164e+06 – 1.203e+061
1.203e+06 – 1.242e+060
1.242e+06 – 1.281e+060
1.281e+06 – 1.319e+060
1.319e+06 – 1.358e+060
1.358e+06 – 1.397e+060
1.397e+06 – 1.436e+060
1.436e+06 – 1.475e+060
1.475e+06 – 1.513e+060
1.513e+06 – 1.552e+061
renter_occupied · The most extreme skew in the dataset (skew ~15.8) — watch for a small number of dense urban counties.
Show data table
bincount
28 – 4.53e+043019
4.53e+04 – 9.057e+04109
9.057e+04 – 1.358e+0538
1.358e+05 – 1.811e+0517
1.811e+05 – 2.264e+0511
2.264e+05 – 2.717e+059
2.717e+05 – 3.169e+055
3.169e+05 – 3.622e+050
3.622e+05 – 4.075e+052
4.075e+05 – 4.528e+052
4.528e+05 – 4.98e+053
4.98e+05 – 5.433e+051
5.433e+05 – 5.886e+051
5.886e+05 – 6.338e+051
6.338e+05 – 6.791e+050
6.791e+05 – 7.244e+051
7.244e+05 – 7.697e+051
7.697e+05 – 8.149e+050
8.149e+05 – 8.602e+050
8.602e+05 – 9.055e+051
9.055e+05 – 9.508e+050
9.508e+05 – 9.96e+050
9.96e+05 – 1.041e+060
1.041e+06 – 1.087e+060
1.087e+06 – 1.132e+060
1.132e+06 – 1.177e+060
1.177e+06 – 1.222e+060
1.222e+06 – 1.268e+060
1.268e+06 – 1.313e+060
1.313e+06 – 1.358e+060
1.358e+06 – 1.403e+060
1.403e+06 – 1.449e+060
1.449e+06 – 1.494e+060
1.494e+06 – 1.539e+060
1.539e+06 – 1.585e+060
1.585e+06 – 1.63e+060
1.63e+06 – 1.675e+060
1.675e+06 – 1.72e+060
1.72e+06 – 1.766e+060
1.766e+06 – 1.811e+061
fips · Roughly uniform across the FIPS range, confirming nationwide county coverage with no clustering.
Show data table
bincount
1001 – 278097
2780 – 455915
4559 – 6337133
6337 – 811659
8116 – 989514
9895 – 1.167e+044
1.167e+04 – 1.345e+04226
1.345e+04 – 1.523e+045
1.523e+04 – 1.701e+0449
1.701e+04 – 1.879e+04189
1.879e+04 – 2.057e+04204
2.057e+04 – 2.235e+04184
2.235e+04 – 2.413e+0439
2.413e+04 – 2.59e+0415
2.59e+04 – 2.768e+04170
2.768e+04 – 2.946e+04196
2.946e+04 – 3.124e+04150
3.124e+04 – 3.302e+0427
3.302e+04 – 3.48e+0421
3.48e+04 – 3.658e+0495
3.658e+04 – 3.836e+04153
3.836e+04 – 4.013e+04155
4.013e+04 – 4.191e+0446
4.191e+04 – 4.369e+0467
4.369e+04 – 4.547e+0451
4.547e+04 – 4.725e+04161
4.725e+04 – 4.903e+04268
4.903e+04 – 5.081e+0429
5.081e+04 – 5.259e+04133
5.259e+04 – 5.436e+0494
5.436e+04 – 5.614e+0495
5.614e+04 – 5.792e+040
5.792e+04 – 5.97e+040
5.97e+04 – 6.148e+040
6.148e+04 – 6.326e+040
6.326e+04 – 6.504e+040
6.504e+04 – 6.682e+040
6.682e+04 – 6.86e+040
6.86e+04 – 7.037e+040
7.037e+04 – 7.215e+0478

Schema

6 columns
Alerts
fips numeric 0.0% 3,222
county_name text 0.0% 3,222
near_unique
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

fips

numeric identifier
This column is the FIPS code for U.S. counties — every one of 3,222 rows is unique with no nulls, matching the count of U.S. counties. Values span 1001 to 72153, consistent with state-prefixed county FIPS identifiers, and the distribution is essentially uniform across the code space (skew 0.157, kurtosis -0.63, no outliers). Treatment: Treat as a categorical key; left-join on this code to county-level reference data rather than using as a numeric feature. high · anthropic:claude-opus-4-7
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_unique
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, so roughly 223 rows use a different administrative suffix (parish, borough, census area). Texas (256), Virginia (189), and Georgia (159) lead the state distribution, consistent with the real US county count. Treatment: use as a join key after splitting into county and state components. high · anthropic:claude-opus-4-7
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_housing_units

numeric feature high_skew outliers
Counts of total housing units per record, almost certainly at a county or similar geographic level given 3,222 rows with 3,074 unique values and no nulls. The distribution is severely right-skewed (skew 12.05, kurtosis 240.5) with a median of 10,021 but a max of 3,363,093, and 443 rows (13.7%) flagged as outliers well above the Q3 of 25,939. The mean of 39,402 sits far above the median, confirming a long heavy tail driven by a few very large geographies. Treatment: log-transform before modelling to tame the heavy right tail. high · anthropic:claude-opus-4-7
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 outliers
This appears to be a count of owner-occupied housing units per geographic area, with 3001 unique values across 3222 rows and effectively no zeros (zero_rate 0.0003) or nulls. The distribution is severely right-skewed (skew 9.52, kurtosis 146.9): the median is 7325.5 but the mean is 25551.7 and the max reaches 1,552,164, producing 429 outliers (13.3% outlier rate). The interquartile range (3147.75 to 18863.5) is dwarfed by the standard deviation of 67553, indicating a long tail of large jurisdictions. Treatment: Log-transform before modelling to tame the heavy right tail. high · anthropic:claude-opus-4-7
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 outliers
Counts of renter-occupied housing units per record, ranging from 28 to 1,810,929 with a median of 2,579.5 — consistent with a geographic rollup (likely county or similar). The distribution is extremely right-skewed (skew 15.82, kurtosis 398.15) and 13.9% of rows fall outside the IQR fences, reflecting a few very large metros dominating a long tail of small areas. No nulls or zeros, and 2,709 unique values across 3,222 rows. Treatment: log-transform before modelling to tame the skew and heavy outliers. high · anthropic:claude-opus-4-7
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 feature
This is a numeric feature representing the percentage of renters per record, ranging from 3.01 to 100.0 with a mean of 27.35 and median of 26.07. The distribution is right-skewed (skew 1.32, kurtosis 4.41) with 88 outliers (2.7%) on the high end, suggesting a small set of records — likely dense urban areas — with renter shares far above the typical 21.64–31.66 IQR. No nulls or zeros, and 1925 unique values across 3222 rows indicate well-populated continuous data. Treatment: Use as-is or apply a mild transform (e.g., log or winsorize) before regression to dampen the right-skew. high · anthropic:claude-opus-4-7
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