saturn·

← all readings

education education by county

source /home/coolhand/datasets/us-inequality-atlas/education/education_by_county.csv 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 contains 3,222 US county-level education records with 6 columns covering county identifiers (county_name, fips, state) and educational attainment metrics (pct_hs_or_higher, pct_bachelors_or_higher, total_25_plus). The bachelor's degree rate averages 23.5% but ranges from 0% to nearly 79%, suggesting wide regional disparities worth investigating. The total_25_plus population column is heavily skewed (skew=13.5) with 440 outliers and a max of nearly 6.9 million, so any analysis using it should consider log transforms or per-capita normalization. State coverage is fairly even across 52 entries, with TX, GA, and VA contributing the most counties.

citing: row_count · column_count · columns.pct_bachelors_or_higher.stats · columns.pct_hs_or_higher.stats · columns.total_25_plus.stats · columns.state.top_values · columns.county_name.top_values

pct_bachelors_or_higher · Shows the wide spread of college attainment across counties, from 0% to nearly 79%.
Show data table
bincount
0 – 1.9721
1.972 – 3.9440
3.944 – 5.9154
5.915 – 7.8879
7.887 – 9.85932
9.859 – 11.83135
11.83 – 13.8169
13.8 – 15.77317
15.77 – 17.75328
17.75 – 19.72376
19.72 – 21.69345
21.69 – 23.66262
23.66 – 25.63232
25.63 – 27.6189
27.6 – 29.58123
29.58 – 31.55116
31.55 – 33.52118
33.52 – 35.4996
35.49 – 37.4660
37.46 – 39.4468
39.44 – 41.4140
41.41 – 43.3834
43.38 – 45.3534
45.35 – 47.3224
47.32 – 49.2921
49.29 – 51.2719
51.27 – 53.2415
53.24 – 55.2110
55.21 – 57.1811
57.18 – 59.1510
59.15 – 61.129
61.12 – 63.16
63.1 – 65.075
65.07 – 67.041
67.04 – 69.010
69.01 – 70.981
70.98 – 72.950
72.95 – 74.930
74.93 – 76.91
76.9 – 78.871
pct_hs_or_higher · Most counties cluster near 88-89% high school completion, with a long left tail of low-attainment counties.
Show data table
bincount
33.33 – 34.991
34.99 – 36.650
36.65 – 38.310
38.31 – 39.970
39.97 – 41.620
41.62 – 43.280
43.28 – 44.940
44.94 – 46.60
46.6 – 48.260
48.26 – 49.920
49.92 – 51.580
51.58 – 53.240
53.24 – 54.90
54.9 – 56.561
56.56 – 58.221
58.22 – 59.871
59.87 – 61.533
61.53 – 63.193
63.19 – 64.853
64.85 – 66.512
66.51 – 68.176
68.17 – 69.837
69.83 – 71.4915
71.49 – 73.1530
73.15 – 74.8130
74.81 – 76.4646
76.46 – 78.1260
78.12 – 79.7888
79.78 – 81.44131
81.44 – 83.1174
83.1 – 84.76189
84.76 – 86.42256
86.42 – 88.08289
88.08 – 89.74360
89.74 – 91.39429
91.39 – 93.05460
93.05 – 94.71389
94.71 – 96.37192
96.37 – 98.0347
98.03 – 99.699
total_25_plus · Highly skewed population distribution; consider log-scaling to see the bulk of small/medium counties.
Show data table
bincount
50 – 1.728e+052948
1.728e+05 – 3.455e+05135
3.455e+05 – 5.183e+0553
5.183e+05 – 6.91e+0537
6.91e+05 – 8.638e+0513
8.638e+05 – 1.036e+0610
1.036e+06 – 1.209e+067
1.209e+06 – 1.382e+064
1.382e+06 – 1.555e+062
1.555e+06 – 1.727e+065
1.727e+06 – 1.9e+061
1.9e+06 – 2.073e+061
2.073e+06 – 2.246e+061
2.246e+06 – 2.418e+061
2.418e+06 – 2.591e+060
2.591e+06 – 2.764e+060
2.764e+06 – 2.937e+060
2.937e+06 – 3.109e+062
3.109e+06 – 3.282e+060
3.282e+06 – 3.455e+060
3.455e+06 – 3.628e+060
3.628e+06 – 3.8e+061
3.8e+06 – 3.973e+060
3.973e+06 – 4.146e+060
4.146e+06 – 4.319e+060
4.319e+06 – 4.491e+060
4.491e+06 – 4.664e+060
4.664e+06 – 4.837e+060
4.837e+06 – 5.01e+060
5.01e+06 – 5.182e+060
5.182e+06 – 5.355e+060
5.355e+06 – 5.528e+060
5.528e+06 – 5.7e+060
5.7e+06 – 5.873e+060
5.873e+06 – 6.046e+060
6.046e+06 – 6.219e+060
6.219e+06 – 6.391e+060
6.391e+06 – 6.564e+060
6.564e+06 – 6.737e+060
6.737e+06 – 6.91e+061
state · Counties per state — TX, GA, and VA contribute the highest counts.
Show data table
valuecountshare
TX2547.9%
GA1594.9%
VA1334.1%
KY1203.7%
MO1153.6%
KS1053.3%
IL1023.2%
NC1003.1%
IA993.1%
TN952.9%
NE932.9%
IN922.9%
OH882.7%
MN872.7%
MI832.6%
MS822.5%
PR782.4%
OK772.4%
AR752.3%
WI722.2%

Schema

6 columns
Alerts
fips numeric 0.0% 3,222
county_name text 0.0% 1,960
short_text duplicates
state categorical 0.0% 52
total_25_plus numeric 0.0% 3,140
high_skew outliers
pct_hs_or_higher numeric 0.0% 1,612
pct_bachelors_or_higher numeric 0.0% 1,982

fips

numeric identifier
This column holds U.S. FIPS county/state codes: every one of the 3222 rows is unique, no nulls, and values span 1001 to 72153, consistent with state-prefixed county identifiers. The distribution is roughly symmetric (skew 0.16, kurtosis -0.63) with no outliers, which is expected for an enumerated geographic key rather than a measured quantity. Treat the numeric stats as incidental — these are categorical identifiers, not magnitudes. Treatment: Cast to zero-padded string and use as a join key to county/state geographies; do not model as a number. 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 metadata short_text duplicates
This is a US county-level place-name field: 2-word entries averaging 14 characters, with 'County' appearing 2999 times alongside 'Parish' (64, Louisiana) and 'Municipio' (78, Puerto Rico). Duplication is heavy at 39.2% (1262 rows) because common names like Washington County (30), Jefferson County (25), and Franklin County (24) recur across states — so this column is not unique on its own. With 1960 distinct values across 3222 rows, it needs a state qualifier to act as a key. Treatment: Concatenate with a state/territory code before using as a join key; do not treat as unique. high · anthropic:claude-opus-4-7
n
3,222
nulls
0 (0.0%)
unique
1,960
len_min
10
len_max
46
len_mean
14.17
len_median
14
len_p95
18
word_mean
2.083
word_median
2
n_empty
0
n_duplicates
1,262
duplicate_rate
0.3917
vocab_size
1,963
readability_flesch_mean
33.36
emoji_rate
0
url_rate
0
one_word_rate
0
allcaps_rate
0
boilerplate_rate
0

state

categorical feature
This is a US state code column with 52 distinct values, matching the 50 states plus likely DC and a territory. Distribution is fairly even (entropy ratio 0.93), with TX leading at 254 rows (7.88%) followed by GA, VA, and KY — consistent with county-level row counts where larger states contribute more records. No nulls. Treatment: one-hot or target-encode for modelling; usable as a join key to state-level reference tables. high · anthropic:claude-opus-4-7
n
3,222
nulls
0 (0.0%)
unique
52
top_value
TX
top_rate
0.07883
cardinality
52
entropy
5.314
entropy_ratio
0.9322

total_25_plus

numeric feature high_skew outliers
A heavily right-skewed count of population (or units) aged 25 and over per row, ranging from 50 to 6,909,650 with a median of 18,313.5 but a mean of 71,074.3. Skew of 13.5 and kurtosis of 306.9 indicate a long heavy tail, and 440 rows (13.7%) flag as outliers. No nulls or zeros, and 3,140 of 3,222 values are unique, consistent with geographic aggregates of varying size. Treatment: log-transform before any regression or distance-based modelling. high · anthropic:claude-opus-4-7
n
3,222
nulls
0 (0.0%)
unique
3,140
min
50
max
6.91e+06
mean
7.107e+04
median
1.831e+04
std
2.266e+05
q1
7696
q3
4.649e+04
iqr
3.879e+04
skew
13.51
kurtosis
306.9
n_outliers
440
outlier_rate
0.1366
zero_rate
0

pct_hs_or_higher

numeric feature
This column reports the percentage of a population with at least a high school education, ranging from 33.33 to 99.69 with a mean of 88.08 and median of 89.39. The distribution is left-skewed (skew -1.33) with heavy tails (kurtosis 3.74), and 86 low-end outliers (2.67%) pull below the bulk concentrated between Q1 84.9 and Q3 92.47. With 1612 unique values across 3222 rows and no nulls, it looks like a clean geographic feature (likely county- or tract-level). Treatment: Use as-is or apply a reflected log/Box-Cox to address the left skew before linear modelling. high · anthropic:claude-opus-4-7
n
3,222
nulls
0 (0.0%)
unique
1,612
min
33.33
max
99.69
mean
88.08
median
89.39
std
5.97
q1
84.9
q3
92.47
iqr
7.567
skew
-1.328
kurtosis
3.742
n_outliers
86
outlier_rate
0.02669
zero_rate
0

pct_bachelors_or_higher

numeric feature
This column reports the percentage of residents with a bachelor's degree or higher across 3,222 rows, ranging from 0.0 to 78.87 with a median of 21.07 and mean of 23.50. The distribution is right-skewed (skew 1.36, kurtosis 2.31) with 141 high-end outliers (4.4%) reflecting a long tail of highly-educated areas. Near-zero zero_rate (0.0003) and no nulls suggest clean coverage. Treatment: Consider a log or sqrt transform to tame the right skew before linear modelling. high · anthropic:claude-opus-4-7
n
3,222
nulls
0 (0.0%)
unique
1,982
min
0
max
78.87
mean
23.5
median
21.07
std
9.983
q1
16.59
q3
27.85
iqr
11.26
skew
1.357
kurtosis
2.306
n_outliers
141
outlier_rate
0.04376
zero_rate
0.0003104