Source: /home/coolhand/html/datavis/data_trove/demographic/veterans/military_firearm_va_healthcare.csv
Saturn profiled 50 rows across 2 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).
!pip install saturn-dissect
import subprocess
subprocess.run([
"saturn", "analyze", "/home/coolhand/html/datavis/data_trove/demographic/veterans/military_firearm_va_healthcare.csv",
"--findings", "data-trove-veteran-homelessness.json",
"--llm", "anthropic:default",
])This dataset contains one row per U.S. state (all 50, no nulls) with a single metric: the percentage of veterans utilizing VA healthcare. The utilization rate ranges from 13.8% to 42.3%, with a mean and median both near 27%, suggesting a roughly symmetric distribution across states. The wide spread — an IQR of about 13 percentage points and a standard deviation of ~8 points — means some states have nearly triple the VA uptake of others, which is worth investigating. Identifying which states cluster at the high and low ends could reveal geographic, demographic, or access-related patterns driving VA healthcare engagement.
citing: row_count · column_count · stats.min · stats.max · stats.mean · stats.median · stats.iqr · stats.std · n_unique
saturn.schema() · 2 columns
| column | kind | n | null% | unique | alerts |
|---|---|---|---|---|---|
| state | categorical | 50 | 0.0% | 50 | long_tail |
| va_utilization_pct | numeric | 50 | 0.0% | 50 |
| bin | count |
|---|---|
| 13.8 – 17.87 | 8 |
| 17.87 – 21.94 | 7 |
| 21.94 – 26.01 | 8 |
| 26.01 – 30.09 | 7 |
| 30.09 – 34.16 | 8 |
| 34.16 – 38.23 | 7 |
| 38.23 – 42.3 | 5 |
| value | count | share |
|---|---|---|
| Alaska | 1 | 2.0% |
| Hawaii | 1 | 2.0% |
| Montana | 1 | 2.0% |
| Wyoming | 1 | 2.0% |
| Vermont | 1 | 2.0% |
| West Virginia | 1 | 2.0% |
| Maine | 1 | 2.0% |
| New Mexico | 1 | 2.0% |
| South Dakota | 1 | 2.0% |
| North Dakota | 1 | 2.0% |
| Oklahoma | 1 | 2.0% |
| Arkansas | 1 | 2.0% |
| Kentucky | 1 | 2.0% |
| Mississippi | 1 | 2.0% |
| Alabama | 1 | 2.0% |
| Tennessee | 1 | 2.0% |
| Louisiana | 1 | 2.0% |
| South Carolina | 1 | 2.0% |
| Missouri | 1 | 2.0% |
| Kansas | 1 | 2.0% |
| bin | count |
|---|---|
| 13.8 – 17.87 | 8 |
| 17.87 – 21.94 | 7 |
| 21.94 – 26.01 | 8 |
| 26.01 – 30.09 | 7 |
| 30.09 – 34.16 | 8 |
| 34.16 – 38.23 | 7 |
| 38.23 – 42.3 | 5 |
| column | kind | null % |
|---|---|---|
| state | categorical | 0.0% |
| va_utilization_pct | numeric | 0.0% |
This column contains U.S. state names and appears to be a reference or lookup table with exactly one row per state — all 50 U.S. states, each appearing exactly once. Cardinality equals n (50), entropy_ratio is exactly 1.0, and top_rate is 0.02 (1/50), confirming perfectly uniform distribution with zero duplication. The 'long_tail' alert is a false positive here: the distribution is maximally flat, not skewed.
Treatment: Use as a join key or dimension label; one-hot encoding would produce 50 sparse columns — prefer target encoding or a regional grouping instead.
saturn.columns["state"].stats
| stat | value |
|---|---|
| n | 50 |
| nulls | 0 (0.0%) |
| unique | 50 |
| top_value | Alaska |
| top_rate | 0.02 |
| cardinality | 50 |
| entropy | 5.644 |
| entropy_ratio | 1 |
| alert: long_tail | 50 singleton categories |
state.
| value | count | share |
|---|---|---|
| Alaska | 1 | 2.0% |
| Hawaii | 1 | 2.0% |
| Montana | 1 | 2.0% |
| Wyoming | 1 | 2.0% |
| Vermont | 1 | 2.0% |
| West Virginia | 1 | 2.0% |
| Maine | 1 | 2.0% |
| New Mexico | 1 | 2.0% |
| South Dakota | 1 | 2.0% |
| North Dakota | 1 | 2.0% |
| Oklahoma | 1 | 2.0% |
| Arkansas | 1 | 2.0% |
| Kentucky | 1 | 2.0% |
| Mississippi | 1 | 2.0% |
| Alabama | 1 | 2.0% |
| Tennessee | 1 | 2.0% |
| Louisiana | 1 | 2.0% |
| South Carolina | 1 | 2.0% |
| Missouri | 1 | 2.0% |
| Kansas | 1 | 2.0% |
This column represents a VA (Veterans Affairs) utilization percentage, likely measuring the share of eligible veterans using VA healthcare or benefits across 50 geographic or administrative units. The distribution is notably uniform and well-behaved: nearly symmetric (skew 0.057), platykurtic (kurtosis -1.12), no outliers, and all 50 values are unique across a tight range of 13.8–42.3%. The near-identical mean (27.38) and median (27.3) with a moderate IQR of 13.43 points suggest consistent utilization rates without extreme variation — unusual regularity for a real-world utilization metric.
Treatment: Use as-is for regression or benchmarking; platykurtic distribution supports linear models without transformation.
saturn.columns["va_utilization_pct"].stats
| stat | value |
|---|---|
| n | 50 |
| nulls | 0 (0.0%) |
| unique | 50 |
| min | 13.8 |
| max | 42.3 |
| mean | 27.38 |
| median | 27.3 |
| std | 8.156 |
| q1 | 20.55 |
| q3 | 33.98 |
| iqr | 13.43 |
| skew | 0.05719 |
| kurtosis | -1.125 |
| n_outliers | 0 |
| outlier_rate | 0 |
| zero_rate | 0 |
va_utilization_pct. Vertical dash marks the median.
| bin | count |
|---|---|
| 13.8 – 17.87 | 8 |
| 17.87 – 21.94 | 7 |
| 21.94 – 26.01 | 8 |
| 26.01 – 30.09 | 7 |
| 30.09 – 34.16 | 8 |
| 34.16 – 38.23 | 7 |
| 38.23 – 42.3 | 5 |
click to copy
@misc{saturn-data-trove-veteran-homelessness-2026,
author = {Steuber, Luke},
title = {Saturn reading: data trove veteran homelessness},
year ={2026},
howpublished = {\url{https://dr.eamer.dev/saturn/view/data-trove-veteran-homelessness}},
note = {Profiled with saturn-dissect v0.2.0, prompt saturn-insight-v2, model anthropic:default},
}
Steuber, L. (2026). Saturn reading: data trove veteran homelessness. Source: /home/coolhand/html/datavis/data_trove/demographic/veterans/military_firearm_va_healthcare.csv. Profiled with saturn-dissect v0.2.0 (saturn-insight-v2, anthropic:default). Retrieved from https://dr.eamer.dev/saturn/view/data-trove-veteran-homelessness