food_deserts-vehicle_access · saturn notebook

Overview

Source: /home/coolhand/html/datavis/data_trove/data/urban/food_deserts/vehicle_access.csv

Saturn profiled 3,222 rows across 9 columns. The stats below are deterministic and machine-readable; the prose is a language-model interpretation of those stats (opt-in, added after the fact, never sees raw rows).

[2]:

!pip install saturn-dissect
import subprocess
subprocess.run([
    "saturn", "analyze", "/home/coolhand/html/datavis/data_trove/data/urban/food_deserts/vehicle_access.csv",
    "--findings", "food_deserts-vehicle_access.json",
    "--llm", "anthropic:claude-opus-4-7",
])

Summary confidence: high

This dataset covers vehicle access for 3,222 US counties (one row per county, identified by FIPS code and name) across 9 columns, with no missing values. The household and no-vehicle counts are extremely right-skewed — `no_vehicle_total` has a median of 580 but a max of 601,621, and `total_households` ranges from 32 up to roughly 3.36 million — so a handful of large urban counties dominate the absolute totals. The more comparable signal is `no_vehicle_pct`, which has a median of 5.41% but stretches up to 85.94%, flagging a small set of counties with extreme transit dependence worth investigating first. State coverage looks complete (52 distinct state codes), so geographic breakdowns should be straightforward.

citing: row_count · column_count · columns.no_vehicle_total.stats · columns.no_vehicle_pct.stats · columns.total_households.stats · columns.state.n_unique · columns.name.n_unique · columns.fips.n_unique

Out[4]:

saturn.schema() · 9 columns

column	kind	n	null%	unique	alerts
name	text	3,222	0.0%	3,222	near_unique
total_households	numeric	3,222	0.0%	3,074	high_skew outliers
no_vehicle_owner	numeric	3,222	0.0%	1,176	high_skew outliers
no_vehicle_renter	numeric	3,222	0.0%	1,517	high_skew outliers
state	numeric	3,222	0.0%	52
county	numeric	3,222	0.0%	330	high_skew outliers
fips	numeric	3,222	0.0%	3,222
no_vehicle_total	numeric	3,222	0.0%	1,823	high_skew outliers
no_vehicle_pct	numeric	3,222	0.0%	1,065	high_skew

Fig 1.

no_vehicle_pct · Distribution of the share of households without a vehicle per county — watch the long right tail above ~20%.

Show data table

Histogram bins for no_vehicle_pct (median: 5.41).
bin	count
0 – 2.148	161
2.148 – 4.297	823
4.297 – 6.445	1091
6.445 – 8.594	630
8.594 – 10.74	283
10.74 – 12.89	111
12.89 – 15.04	61
15.04 – 17.19	23
17.19 – 19.34	8
19.34 – 21.48	3
21.48 – 23.63	4
23.63 – 25.78	2
25.78 – 27.93	3
27.93 – 30.08	2
30.08 – 32.23	2
32.23 – 34.38	2
34.38 – 36.52	2
36.52 – 38.67	2
38.67 – 40.82	0
40.82 – 42.97	1
42.97 – 45.12	0
45.12 – 47.27	1
47.27 – 49.42	0
49.42 – 51.56	0
51.56 – 53.71	0
53.71 – 55.86	1
55.86 – 58.01	1
58.01 – 60.16	0
60.16 – 62.31	2
62.31 – 64.45	0
64.45 – 66.6	1
66.6 – 68.75	0
68.75 – 70.9	0
70.9 – 73.05	0
73.05 – 75.2	0
75.2 – 77.35	0
77.35 – 79.49	1
79.49 – 81.64	0
81.64 – 83.79	0
83.79 – 85.94	1

Fig 2.

no_vehicle_total · Absolute count of no-vehicle households is extremely skewed; a few counties exceed 100k while the median is just 580.

Show data table

Histogram bins for no_vehicle_total (median: 580.0).
bin	count
0 – 1.504e+04	3108
1.504e+04 – 3.008e+04	62
3.008e+04 – 4.512e+04	19
4.512e+04 – 6.016e+04	9
6.016e+04 – 7.52e+04	8
7.52e+04 – 9.024e+04	2
9.024e+04 – 1.053e+05	4
1.053e+05 – 1.203e+05	3
1.203e+05 – 1.354e+05	0
1.354e+05 – 1.504e+05	0
1.504e+05 – 1.654e+05	0
1.654e+05 – 1.805e+05	0
1.805e+05 – 1.955e+05	1
1.955e+05 – 2.106e+05	0
2.106e+05 – 2.256e+05	0
2.256e+05 – 2.406e+05	0
2.406e+05 – 2.557e+05	0
2.557e+05 – 2.707e+05	0
2.707e+05 – 2.858e+05	0
2.858e+05 – 3.008e+05	2
3.008e+05 – 3.159e+05	0
3.159e+05 – 3.309e+05	1
3.309e+05 – 3.459e+05	0
3.459e+05 – 3.61e+05	0
3.61e+05 – 3.76e+05	1
3.76e+05 – 3.911e+05	0
3.911e+05 – 4.061e+05	0
4.061e+05 – 4.211e+05	0
4.211e+05 – 4.362e+05	0
4.362e+05 – 4.512e+05	0
4.512e+05 – 4.663e+05	0
4.663e+05 – 4.813e+05	0
4.813e+05 – 4.963e+05	0
4.963e+05 – 5.114e+05	0
5.114e+05 – 5.264e+05	0
5.264e+05 – 5.415e+05	0
5.415e+05 – 5.565e+05	1
5.565e+05 – 5.715e+05	0
5.715e+05 – 5.866e+05	0
5.866e+05 – 6.016e+05	1

Fig 3.

total_households · County household sizes span five orders of magnitude — useful context before comparing raw no-vehicle totals.

Show data table

Histogram bins for total_households (median: 10021.0).
bin	count
32 – 8.411e+04	2907
8.411e+04 – 1.682e+05	153
1.682e+05 – 2.523e+05	62
2.523e+05 – 3.363e+05	38
3.363e+05 – 4.204e+05	22
4.204e+05 – 5.045e+05	6
5.045e+05 – 5.886e+05	11
5.886e+05 – 6.726e+05	5
6.726e+05 – 7.567e+05	5
7.567e+05 – 8.408e+05	3
8.408e+05 – 9.249e+05	1
9.249e+05 – 1.009e+06	3
1.009e+06 – 1.093e+06	1
1.093e+06 – 1.177e+06	1
1.177e+06 – 1.261e+06	0
1.261e+06 – 1.345e+06	0
1.345e+06 – 1.429e+06	0
1.429e+06 – 1.513e+06	0
1.513e+06 – 1.597e+06	0
1.597e+06 – 1.682e+06	1
1.682e+06 – 1.766e+06	1
1.766e+06 – 1.85e+06	0
1.85e+06 – 1.934e+06	0
1.934e+06 – 2.018e+06	0
2.018e+06 – 2.102e+06	1
2.102e+06 – 2.186e+06	0
2.186e+06 – 2.27e+06	0
2.27e+06 – 2.354e+06	0
2.354e+06 – 2.438e+06	0
2.438e+06 – 2.522e+06	0
2.522e+06 – 2.606e+06	0
2.606e+06 – 2.69e+06	0
2.69e+06 – 2.775e+06	0
2.775e+06 – 2.859e+06	0
2.859e+06 – 2.943e+06	0
2.943e+06 – 3.027e+06	0
3.027e+06 – 3.111e+06	0
3.111e+06 – 3.195e+06	0
3.195e+06 – 3.279e+06	0
3.279e+06 – 3.363e+06	1

Fig 4.

state · Counties per state code — confirms geographic coverage across all 52 state-level FIPS values.

Show data table

Histogram bins for state (median: 30.0).
bin	count
1 – 2.775	97
2.775 – 4.55	15
4.55 – 6.325	133
6.325 – 8.1	64
8.1 – 9.875	9
9.875 – 11.65	4
11.65 – 13.42	226
13.42 – 15.2	5
15.2 – 16.98	44
16.98 – 18.75	194
18.75 – 20.52	204
20.52 – 22.3	184
22.3 – 24.07	40
24.07 – 25.85	14
25.85 – 27.62	170
27.62 – 29.4	197
29.4 – 31.17	149
31.17 – 32.95	17
32.95 – 34.73	31
34.73 – 36.5	95
36.5 – 38.27	153
38.27 – 40.05	165
40.05 – 41.82	36
41.82 – 43.6	67
43.6 – 45.38	51
45.38 – 47.15	161
47.15 – 48.92	254
48.92 – 50.7	43
50.7 – 52.47	133
52.47 – 54.25	94
54.25 – 56.02	95
56.02 – 57.8	0
57.8 – 59.57	0
59.57 – 61.35	0
61.35 – 63.12	0
63.12 – 64.9	0
64.9 – 66.67	0
66.67 – 68.45	0
68.45 – 70.22	0
70.22 – 72	78

Fig 5.

no_vehicle_renter · Renter households without vehicles skew even harder than owners; helpful for spotting urban renter-heavy counties.

Show data table

Histogram bins for no_vehicle_renter (median: 351.0).
bin	count
0 – 1.22e+04	3112
1.22e+04 – 2.441e+04	63
2.441e+04 – 3.661e+04	16
3.661e+04 – 4.881e+04	8
4.881e+04 – 6.102e+04	8
6.102e+04 – 7.322e+04	2
7.322e+04 – 8.543e+04	1
8.543e+04 – 9.763e+04	5
9.763e+04 – 1.098e+05	0
1.098e+05 – 1.22e+05	0
1.22e+05 – 1.342e+05	1
1.342e+05 – 1.464e+05	0
1.464e+05 – 1.586e+05	0
1.586e+05 – 1.709e+05	0
1.709e+05 – 1.831e+05	0
1.831e+05 – 1.953e+05	0
1.953e+05 – 2.075e+05	0
2.075e+05 – 2.197e+05	0
2.197e+05 – 2.319e+05	1
2.319e+05 – 2.441e+05	1
2.441e+05 – 2.563e+05	0
2.563e+05 – 2.685e+05	0
2.685e+05 – 2.807e+05	0
2.807e+05 – 2.929e+05	2
2.929e+05 – 3.051e+05	0
3.051e+05 – 3.173e+05	0
3.173e+05 – 3.295e+05	0
3.295e+05 – 3.417e+05	0
3.417e+05 – 3.539e+05	0
3.539e+05 – 3.661e+05	0
3.661e+05 – 3.783e+05	0
3.783e+05 – 3.905e+05	0
3.905e+05 – 4.027e+05	0
4.027e+05 – 4.149e+05	0
4.149e+05 – 4.271e+05	0
4.271e+05 – 4.393e+05	0
4.393e+05 – 4.515e+05	0
4.515e+05 – 4.637e+05	0
4.637e+05 – 4.759e+05	1
4.759e+05 – 4.881e+05	1

Fig 6.

Per-column null rate across the corpus. Columns are ordered by input position.

Show data table

Per-column null rate across the corpus.
column	kind	null %
name	text	0.0%
total_households	numeric	0.0%
no_vehicle_owner	numeric	0.0%
no_vehicle_renter	numeric	0.0%
state	numeric	0.0%
county	numeric	0.0%
fips	numeric	0.0%
no_vehicle_total	numeric	0.0%
no_vehicle_pct	numeric	0.0%

Fig 7.

Pearson correlation across numeric columns (sampled, bounded).

Show data table

Pearson correlation across 8 numeric columns (values clipped to 2 decimals).
	total_households	no_vehicle_owner	no_vehicle_renter	state	county	fips	no_vehicle_total	no_vehicle_pct
total_households	+1.00	+0.92	+0.90	-0.06	-0.02	-0.06	+0.91	+0.10
no_vehicle_owner	+0.92	+1.00	+0.97	-0.05	-0.03	-0.06	+0.98	+0.16
no_vehicle_renter	+0.90	+0.97	+1.00	-0.05	-0.02	-0.05	+1.00	+0.15
state	-0.06	-0.05	-0.05	+1.00	+0.14	+1.00	-0.05	+0.04
county	-0.02	-0.03	-0.02	+0.14	+1.00	+0.15	-0.02	+0.11
fips	-0.06	-0.06	-0.05	+1.00	+0.15	+1.00	-0.05	+0.04
no_vehicle_total	+0.91	+0.98	+1.00	-0.05	-0.02	-0.05	+1.00	+0.15
no_vehicle_pct	+0.10	+0.16	+0.15	+0.04	+0.11	+0.04	+0.15	+1.00

name text identifier

This column appears to hold US county names with state suffixes — 2,999 of 3,222 rows contain the token 'county,' and the remaining top words are state names (texas, virginia, georgia, north carolina, dakota, kentucky, missouri). Every value is unique (n_unique=3222, duplicate_rate=0.0) with no nulls, and lengths are tightly clustered (mean 24.3, min 16, max 59, p95 31), consistent with 'X County, State' formatting. The near_unique alert confirms this behaves as a row identifier rather than a categorical feature.

Treatment: Use as a row label or join key on county; do not one-hot encode.

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

Out[13]:

saturn.columns["name"].stats

stat	value
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
alert: near_unique	100.0% of rows are unique strings

Fig 8.

Character-length distribution for name.

Show data table

Character-length distribution for name (mean: 24.324022346368714).
chars	count
16 – 17	26
17 – 18	72
18 – 19	121
19 – 20	190
20 – 21	264
21 – 22	407
22 – 24	420
24 – 25	363
25 – 26	320
26 – 27	240
27 – 28	231
28 – 29	152
29 – 30	139
30 – 31	165
31 – 32	41
32 – 33	28
33 – 34	16
34 – 35	10
35 – 36	5
36 – 38	0
38 – 39	1
39 – 40	1
40 – 41	0
41 – 42	1
42 – 43	1
43 – 44	0
44 – 45	2
45 – 46	0
46 – 47	1
47 – 48	1
48 – 49	0
49 – 50	0
50 – 51	0
51 – 53	0
53 – 54	2
54 – 55	1
55 – 56	0
56 – 57	0
57 – 58	0
58 – 59	1

total_households numeric feature

Likely a count of households per geographic unit (e.g., county or tract), with 3222 rows, 3074 unique values, and no nulls or zeros. The distribution is extremely right-skewed (skew 12.05, kurtosis 240.5): the median is 10021 while the mean is 39402.86 and the max reaches 3363093, roughly 28x the standard deviation above the mean. Saturn flags 443 outliers (13.7%), consistent with a few very large jurisdictions dominating the tail.

Treatment: Log-transform before regression to tame the skew and outliers.

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

Out[16]:

saturn.columns["total_households"].stats

stat	value
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
alert: high_skew	skew=+12.05
alert: outliers	13.7% rows beyond 1.5 IQR

Fig 9.

Distribution of total_households. Vertical dash marks the median.

Show data table

Histogram bins for total_households (median: 10021.0).
bin	count
32 – 8.411e+04	2907
8.411e+04 – 1.682e+05	153
1.682e+05 – 2.523e+05	62
2.523e+05 – 3.363e+05	38
3.363e+05 – 4.204e+05	22
4.204e+05 – 5.045e+05	6
5.045e+05 – 5.886e+05	11
5.886e+05 – 6.726e+05	5
6.726e+05 – 7.567e+05	5
7.567e+05 – 8.408e+05	3
8.408e+05 – 9.249e+05	1
9.249e+05 – 1.009e+06	3
1.009e+06 – 1.093e+06	1
1.093e+06 – 1.177e+06	1
1.177e+06 – 1.261e+06	0
1.261e+06 – 1.345e+06	0
1.345e+06 – 1.429e+06	0
1.429e+06 – 1.513e+06	0
1.513e+06 – 1.597e+06	0
1.597e+06 – 1.682e+06	1
1.682e+06 – 1.766e+06	1
1.766e+06 – 1.85e+06	0
1.85e+06 – 1.934e+06	0
1.934e+06 – 2.018e+06	0
2.018e+06 – 2.102e+06	1
2.102e+06 – 2.186e+06	0
2.186e+06 – 2.27e+06	0
2.27e+06 – 2.354e+06	0
2.354e+06 – 2.438e+06	0
2.438e+06 – 2.522e+06	0
2.522e+06 – 2.606e+06	0
2.606e+06 – 2.69e+06	0
2.69e+06 – 2.775e+06	0
2.775e+06 – 2.859e+06	0
2.859e+06 – 2.943e+06	0
2.943e+06 – 3.027e+06	0
3.027e+06 – 3.111e+06	0
3.111e+06 – 3.195e+06	0
3.195e+06 – 3.279e+06	0
3.279e+06 – 3.363e+06	1

no_vehicle_owner numeric feature

Likely a count of non-vehicle-owners aggregated per row (e.g., per geography or unit), ranging from 0 to 113,473 with a median of just 214. The distribution is severely right-skewed (skew 18.55, kurtosis 433.5) with 360 outliers (11.2%) and a std (3777.8) nearly 5x the mean (820.8), signalling a heavy tail dominated by a few extreme rows. Only 1.2% of rows are zero and there are no nulls, so the column is densely populated but dispersed across 1,176 unique values.

Treatment: Apply a log1p transform before modelling to tame the heavy right tail.

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

Out[19]:

saturn.columns["no_vehicle_owner"].stats

stat	value
n	3,222
nulls	0 (0.0%)
unique	1,176
min	0
max	113,473
mean	820.8
median	214
std	3778
q1	81
q3	548.8
iqr	467.8
skew	18.55
kurtosis	433.5
n_outliers	360
outlier_rate	0.1117
zero_rate	0.01179
alert: high_skew	skew=+18.55
alert: outliers	11.2% rows beyond 1.5 IQR

Fig 10.

Distribution of no_vehicle_owner. Vertical dash marks the median.

Show data table

Histogram bins for no_vehicle_owner (median: 214.0).
bin	count
0 – 2837	3068
2837 – 5674	82
5674 – 8510	33
8510 – 1.135e+04	10
1.135e+04 – 1.418e+04	13
1.418e+04 – 1.702e+04	3
1.702e+04 – 1.986e+04	2
1.986e+04 – 2.269e+04	2
2.269e+04 – 2.553e+04	2
2.553e+04 – 2.837e+04	1
2.837e+04 – 3.121e+04	0
3.121e+04 – 3.404e+04	0
3.404e+04 – 3.688e+04	0
3.688e+04 – 3.972e+04	0
3.972e+04 – 4.255e+04	0
4.255e+04 – 4.539e+04	0
4.539e+04 – 4.823e+04	1
4.823e+04 – 5.106e+04	0
5.106e+04 – 5.39e+04	0
5.39e+04 – 5.674e+04	0
5.674e+04 – 5.957e+04	1
5.957e+04 – 6.241e+04	0
6.241e+04 – 6.525e+04	0
6.525e+04 – 6.808e+04	0
6.808e+04 – 7.092e+04	1
7.092e+04 – 7.376e+04	0
7.376e+04 – 7.659e+04	0
7.659e+04 – 7.943e+04	1
7.943e+04 – 8.227e+04	0
8.227e+04 – 8.51e+04	1
8.51e+04 – 8.794e+04	0
8.794e+04 – 9.078e+04	0
9.078e+04 – 9.362e+04	0
9.362e+04 – 9.645e+04	0
9.645e+04 – 9.929e+04	0
9.929e+04 – 1.021e+05	0
1.021e+05 – 1.05e+05	0
1.05e+05 – 1.078e+05	0
1.078e+05 – 1.106e+05	0
1.106e+05 – 1.135e+05	1

no_vehicle_renter numeric feature

A heavily right-skewed numeric count, plausibly the number of renters without a vehicle per record/area. The median is 351 but the mean is 2483 and the max reaches 488148, with skew 20.7 and kurtosis 517.5 driving 436 outliers (13.5%). About 1.5% of rows are zero and there are no nulls, so the long tail — not missingness — dominates the distribution.

Treatment: Apply a log1p transform before modelling to tame the extreme skew and outliers.

anthropic:claude-opus-4-7 · confidence medium

Out[22]:

saturn.columns["no_vehicle_renter"].stats

stat	value
n	3,222
nulls	0 (0.0%)
unique	1,517
min	0
max	488,148
mean	2483
median	351
std	1.646e+04
q1	125.2
q3	987.8
iqr	862.5
skew	20.69
kurtosis	517.5
n_outliers	436
outlier_rate	0.1353
zero_rate	0.01459
alert: high_skew	skew=+20.69
alert: outliers	13.5% rows beyond 1.5 IQR

Fig 11.

Distribution of no_vehicle_renter. Vertical dash marks the median.

Show data table

Histogram bins for no_vehicle_renter (median: 351.0).
bin	count
0 – 1.22e+04	3112
1.22e+04 – 2.441e+04	63
2.441e+04 – 3.661e+04	16
3.661e+04 – 4.881e+04	8
4.881e+04 – 6.102e+04	8
6.102e+04 – 7.322e+04	2
7.322e+04 – 8.543e+04	1
8.543e+04 – 9.763e+04	5
9.763e+04 – 1.098e+05	0
1.098e+05 – 1.22e+05	0
1.22e+05 – 1.342e+05	1
1.342e+05 – 1.464e+05	0
1.464e+05 – 1.586e+05	0
1.586e+05 – 1.709e+05	0
1.709e+05 – 1.831e+05	0
1.831e+05 – 1.953e+05	0
1.953e+05 – 2.075e+05	0
2.075e+05 – 2.197e+05	0
2.197e+05 – 2.319e+05	1
2.319e+05 – 2.441e+05	1
2.441e+05 – 2.563e+05	0
2.563e+05 – 2.685e+05	0
2.685e+05 – 2.807e+05	0
2.807e+05 – 2.929e+05	2
2.929e+05 – 3.051e+05	0
3.051e+05 – 3.173e+05	0
3.173e+05 – 3.295e+05	0
3.295e+05 – 3.417e+05	0
3.417e+05 – 3.539e+05	0
3.539e+05 – 3.661e+05	0
3.661e+05 – 3.783e+05	0
3.783e+05 – 3.905e+05	0
3.905e+05 – 4.027e+05	0
4.027e+05 – 4.149e+05	0
4.149e+05 – 4.271e+05	0
4.271e+05 – 4.393e+05	0
4.393e+05 – 4.515e+05	0
4.515e+05 – 4.637e+05	0
4.637e+05 – 4.759e+05	1
4.759e+05 – 4.881e+05	1

state numeric feature

Despite being typed as numeric, this column holds 52 distinct integer values between 1 and 72 with no nulls or zeros, which matches a FIPS-style state code encoding (50 states plus DC and territories) rather than a true measurement. The near-uniform spread (mean 31.27, median 30, std 16.29, skew 0.16) and absence of outliers reinforce that these are categorical identifiers, not quantities.

Treatment: Cast to categorical and one-hot or target-encode; do not use as a continuous variable.

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

Out[25]:

saturn.columns["state"].stats

stat	value
n	3,222
nulls	0 (0.0%)
unique	52
min	1
max	72
mean	31.27
median	30
std	16.29
q1	19
q3	46
iqr	27
skew	0.1574
kurtosis	-0.6267
n_outliers	0
outlier_rate	0
zero_rate	0

Fig 12.

Distribution of state. Vertical dash marks the median.

Show data table

Histogram bins for state (median: 30.0).
bin	count
1 – 2.775	97
2.775 – 4.55	15
4.55 – 6.325	133
6.325 – 8.1	64
8.1 – 9.875	9
9.875 – 11.65	4
11.65 – 13.42	226
13.42 – 15.2	5
15.2 – 16.98	44
16.98 – 18.75	194
18.75 – 20.52	204
20.52 – 22.3	184
22.3 – 24.07	40
24.07 – 25.85	14
25.85 – 27.62	170
27.62 – 29.4	197
29.4 – 31.17	149
31.17 – 32.95	17
32.95 – 34.73	31
34.73 – 36.5	95
36.5 – 38.27	153
38.27 – 40.05	165
40.05 – 41.82	36
41.82 – 43.6	67
43.6 – 45.38	51
45.38 – 47.15	161
47.15 – 48.92	254
48.92 – 50.7	43
50.7 – 52.47	133
52.47 – 54.25	94
54.25 – 56.02	95
56.02 – 57.8	0
57.8 – 59.57	0
59.57 – 61.35	0
61.35 – 63.12	0
63.12 – 64.9	0
64.9 – 66.67	0
66.67 – 68.45	0
68.45 – 70.22	0
70.22 – 72	78

county numeric identifier

Despite the name 'county', the column is stored numerically with 330 unique values across 3222 rows and no nulls, suggesting it holds county FIPS codes or similar integer identifiers rather than a measured quantity. The distribution is heavily right-skewed (skew 2.87, kurtosis 11.6) with a median of 79 but a max of 840 and 178 outliers (5.5%), which is expected for code-like values but misleading if treated as a continuous feature.

Treatment: Treat as a categorical code (e.g., FIPS) and one-hot or target-encode rather than using as a numeric feature.

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

Out[28]:

saturn.columns["county"].stats

stat	value
n	3,222
nulls	0 (0.0%)
unique	330
min	1
max	840
mean	103.2
median	79
std	106.6
q1	35
q3	133
iqr	98
skew	2.866
kurtosis	11.64
n_outliers	178
outlier_rate	0.05525
zero_rate	0
alert: high_skew	skew=+2.87
alert: outliers	5.5% rows beyond 1.5 IQR

Fig 13.

Distribution of county. Vertical dash marks the median.

Show data table

Histogram bins for county (median: 79.0).
bin	count
1 – 21.98	523
21.98 – 42.95	418
42.95 – 63.93	411
63.93 – 84.9	345
84.9 – 105.9	352
105.9 – 126.9	281
126.9 – 147.8	236
147.8 – 168.8	168
168.8 – 189.8	140
189.8 – 210.8	71
210.8 – 231.7	45
231.7 – 252.7	25
252.7 – 273.7	22
273.7 – 294.7	23
294.7 – 315.6	22
315.6 – 336.6	13
336.6 – 357.6	11
357.6 – 378.6	10
378.6 – 399.5	11
399.5 – 420.5	10
420.5 – 441.5	11
441.5 – 462.5	10
462.5 – 483.4	11
483.4 – 504.4	10
504.4 – 525.4	7
525.4 – 546.4	2
546.4 – 567.3	1
567.3 – 588.3	2
588.3 – 609.3	3
609.3 – 630.2	3
630.2 – 651.2	2
651.2 – 672.2	2
672.2 – 693.2	5
693.2 – 714.2	2
714.2 – 735.1	3
735.1 – 756.1	2
756.1 – 777.1	3
777.1 – 798.1	1
798.1 – 819	2
819 – 840	3

fips numeric identifier

This is the FIPS county code: every one of the 3222 rows is unique with no nulls, and the value range (1001 to 72153) matches the standard 5-digit state+county FIPS encoding. Distribution stats (mean 31377, median 30022, near-zero skew) are essentially meaningless here since the codes are categorical identifiers, not quantities. No outliers flagged, consistent with a clean geographic key.

Treatment: Treat as a categorical geographic key; left-join on this code rather than using it as a numeric feature.

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

Out[31]:

saturn.columns["fips"].stats

stat	value
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

Fig 14.

Distribution of fips. Vertical dash marks the median.

Show data table

Histogram bins for fips (median: 30022.0).
bin	count
1001 – 2780	97
2780 – 4559	15
4559 – 6337	133
6337 – 8116	59
8116 – 9895	14
9895 – 1.167e+04	4
1.167e+04 – 1.345e+04	226
1.345e+04 – 1.523e+04	5
1.523e+04 – 1.701e+04	49
1.701e+04 – 1.879e+04	189
1.879e+04 – 2.057e+04	204
2.057e+04 – 2.235e+04	184
2.235e+04 – 2.413e+04	39
2.413e+04 – 2.59e+04	15
2.59e+04 – 2.768e+04	170
2.768e+04 – 2.946e+04	196
2.946e+04 – 3.124e+04	150
3.124e+04 – 3.302e+04	27
3.302e+04 – 3.48e+04	21
3.48e+04 – 3.658e+04	95
3.658e+04 – 3.836e+04	153
3.836e+04 – 4.013e+04	155
4.013e+04 – 4.191e+04	46
4.191e+04 – 4.369e+04	67
4.369e+04 – 4.547e+04	51
4.547e+04 – 4.725e+04	161
4.725e+04 – 4.903e+04	268
4.903e+04 – 5.081e+04	29
5.081e+04 – 5.259e+04	133
5.259e+04 – 5.436e+04	94
5.436e+04 – 5.614e+04	95
5.614e+04 – 5.792e+04	0
5.792e+04 – 5.97e+04	0
5.97e+04 – 6.148e+04	0
6.148e+04 – 6.326e+04	0
6.326e+04 – 6.504e+04	0
6.504e+04 – 6.682e+04	0
6.682e+04 – 6.86e+04	0
6.86e+04 – 7.037e+04	0
7.037e+04 – 7.215e+04	78

no_vehicle_total numeric feature

Counts of vehicles (totals) per record, ranging from 0 to 601,621 with a median of 580 but a mean of 3,304. The distribution is extremely right-skewed (skew 20.26, kurtosis 501.27) with 407 outliers (12.6%) and a std (20,050) far exceeding the IQR (1,331.75), indicating a few enormous records dominate.

Treatment: Apply a log1p transform and consider winsorising before modelling.

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

Out[34]:

saturn.columns["no_vehicle_total"].stats

stat	value
n	3,222
nulls	0 (0.0%)
unique	1,823
min	0
max	601,621
mean	3304
median	580
std	2.005e+04
q1	223
q3	1555
iqr	1332
skew	20.26
kurtosis	501.3
n_outliers	407
outlier_rate	0.1263
zero_rate	0.003724
alert: high_skew	skew=+20.26
alert: outliers	12.6% rows beyond 1.5 IQR

Fig 15.

Distribution of no_vehicle_total. Vertical dash marks the median.

Show data table

Histogram bins for no_vehicle_total (median: 580.0).
bin	count
0 – 1.504e+04	3108
1.504e+04 – 3.008e+04	62
3.008e+04 – 4.512e+04	19
4.512e+04 – 6.016e+04	9
6.016e+04 – 7.52e+04	8
7.52e+04 – 9.024e+04	2
9.024e+04 – 1.053e+05	4
1.053e+05 – 1.203e+05	3
1.203e+05 – 1.354e+05	0
1.354e+05 – 1.504e+05	0
1.504e+05 – 1.654e+05	0
1.654e+05 – 1.805e+05	0
1.805e+05 – 1.955e+05	1
1.955e+05 – 2.106e+05	0
2.106e+05 – 2.256e+05	0
2.256e+05 – 2.406e+05	0
2.406e+05 – 2.557e+05	0
2.557e+05 – 2.707e+05	0
2.707e+05 – 2.858e+05	0
2.858e+05 – 3.008e+05	2
3.008e+05 – 3.159e+05	0
3.159e+05 – 3.309e+05	1
3.309e+05 – 3.459e+05	0
3.459e+05 – 3.61e+05	0
3.61e+05 – 3.76e+05	1
3.76e+05 – 3.911e+05	0
3.911e+05 – 4.061e+05	0
4.061e+05 – 4.211e+05	0
4.211e+05 – 4.362e+05	0
4.362e+05 – 4.512e+05	0
4.512e+05 – 4.663e+05	0
4.663e+05 – 4.813e+05	0
4.813e+05 – 4.963e+05	0
4.963e+05 – 5.114e+05	0
5.114e+05 – 5.264e+05	0
5.264e+05 – 5.415e+05	0
5.415e+05 – 5.565e+05	1
5.565e+05 – 5.715e+05	0
5.715e+05 – 5.866e+05	0
5.866e+05 – 6.016e+05	1

no_vehicle_pct numeric feature

Likely the percentage of households (or similar units) with no vehicle, recorded for 3,222 rows with no nulls and values bounded between 0.0 and 85.94. The distribution is tightly clustered (median 5.41, IQR 3.38) but extremely right-skewed (skew 6.98, kurtosis 86.23), with 140 outliers (4.35%) pulling the mean to 6.20 well above the median. Only 0.37% are exact zeros, so true absence is rare; the long tail is the surprise here.

Treatment: Apply a log1p or winsorising transform before modelling to tame the heavy right tail.

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

Out[37]:

saturn.columns["no_vehicle_pct"].stats

stat	value
n	3,222
nulls	0 (0.0%)
unique	1,065
min	0
max	85.94
mean	6.197
median	5.41
std	4.538
q1	3.98
q3	7.36
iqr	3.38
skew	6.976
kurtosis	86.23
n_outliers	140
outlier_rate	0.04345
zero_rate	0.003724
alert: high_skew	skew=+6.98

Fig 16.

Distribution of no_vehicle_pct. Vertical dash marks the median.

Show data table

Histogram bins for no_vehicle_pct (median: 5.41).
bin	count
0 – 2.148	161
2.148 – 4.297	823
4.297 – 6.445	1091
6.445 – 8.594	630
8.594 – 10.74	283
10.74 – 12.89	111
12.89 – 15.04	61
15.04 – 17.19	23
17.19 – 19.34	8
19.34 – 21.48	3
21.48 – 23.63	4
23.63 – 25.78	2
25.78 – 27.93	3
27.93 – 30.08	2
30.08 – 32.23	2
32.23 – 34.38	2
34.38 – 36.52	2
36.52 – 38.67	2
38.67 – 40.82	0
40.82 – 42.97	1
42.97 – 45.12	0
45.12 – 47.27	1
47.27 – 49.42	0
49.42 – 51.56	0
51.56 – 53.71	0
53.71 – 55.86	1
55.86 – 58.01	1
58.01 – 60.16	0
60.16 – 62.31	2
62.31 – 64.45	0
64.45 – 66.6	1
66.6 – 68.75	0
68.75 – 70.9	0
70.9 – 73.05	0
73.05 – 75.2	0
75.2 – 77.35	0
77.35 – 79.49	1
79.49 – 81.64	0
81.64 – 83.79	0
83.79 – 85.94	1

food deserts vehicle access

Overview

Summary confidence: high

name text identifier

total_households numeric feature

no_vehicle_owner numeric feature

no_vehicle_renter numeric feature

state numeric feature

county numeric identifier

fips numeric identifier

no_vehicle_total numeric feature

no_vehicle_pct numeric feature

How to cite