data trove veteran suicide rates

saturn notebook · generated 2026-06-22 Report Notebook

Overview

Source: /home/coolhand/html/datavis/data_trove/demographic/veterans/military_firearm_suicide.csv

Saturn profiled 50 rows across 4 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/demographic/veterans/military_firearm_suicide.csv",
    "--findings", "data-trove-veteran-suicide-rates.json",
    "--llm", "anthropic:default",
])

Summary confidence: high

This dataset contains state-level suicide rate statistics for all 50 U.S. states, comparing civilian and veteran populations along with a veteran risk ratio. The most striking signal is the scale of the veteran suicide burden: the mean veteran suicide rate (36.1 per 100k) is roughly double the civilian mean (17.6 per 100k), and the veteran risk ratio ranges from 1.8 to 3.23, meaning veterans are at minimum nearly twice as likely to die by suicide as civilians in every single state. The right-skewed distribution of the veteran risk ratio deserves closer attention — a handful of states show ratios above 2.4, suggesting particularly acute disparities worth investigating.

citing: mean · median · min · max · skew · iqr · std

Out[4]:

saturn.schema() · 4 columns

column	kind	n	null%	unique	alerts
state	categorical	50	0.0%	50	long_tail
veteran_suicide_rate	numeric	50	0.0%	50
civilian_suicide_rate	numeric	50	0.0%	50
veteran_risk_ratio	numeric	50	0.0%	41

Fig 1.

veteran_suicide_rate · Look for the wide spread (24.9–52.3) and mild right skew, showing which states have the highest veteran suicide rates.

Show data table

Histogram bins for veteran_suicide_rate (median: 35.0).
bin	count
24.9 – 28.81	10
28.81 – 32.73	9
32.73 – 36.64	9
36.64 – 40.56	8
40.56 – 44.47	6
44.47 – 48.39	4
48.39 – 52.3	4

Fig 2.

civilian_suicide_rate · Compare the civilian distribution (7.7–28.9) directly against veterans — the entire veteran range sits far above most civilian values.

Show data table

Histogram bins for civilian_suicide_rate (median: 17.5).
bin	count
7.7 – 10.73	8
10.73 – 13.76	8
13.76 – 16.79	7
16.79 – 19.81	8
19.81 – 22.84	7
22.84 – 25.87	7
25.87 – 28.9	5

Fig 3.

veteran_risk_ratio · The right skew here reveals a subset of states where veterans face a disproportionately higher risk — ratios above 2.4 are the outliers to flag.

Show data table

Histogram bins for veteran_risk_ratio (median: 2.0).
bin	count
1.8 – 2.004	25
2.004 – 2.209	7
2.209 – 2.413	6
2.413 – 2.617	4
2.617 – 2.821	3
2.821 – 3.026	2
3.026 – 3.23	3

Fig 4.

state · Rank states by veteran suicide rate or risk ratio to identify geographic clusters of elevated veteran suicide risk.

Show data table

Top values for state (20 unique shown, of 50 total).
value	count	share
Montana	1	2.0%
Wyoming	1	2.0%
Alaska	1	2.0%
New Mexico	1	2.0%
Idaho	1	2.0%
Oklahoma	1	2.0%
Colorado	1	2.0%
South Dakota	1	2.0%
West Virginia	1	2.0%
Arkansas	1	2.0%
Nevada	1	2.0%
Arizona	1	2.0%
Oregon	1	2.0%
Utah	1	2.0%
Kentucky	1	2.0%
Tennessee	1	2.0%
Alabama	1	2.0%
North Dakota	1	2.0%
Missouri	1	2.0%
Kansas	1	2.0%

Fig 5.

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 %
state	categorical	0.0%
veteran_suicide_rate	numeric	0.0%
civilian_suicide_rate	numeric	0.0%
veteran_risk_ratio	numeric	0.0%

Fig 6.

Pearson correlation across numeric columns (sampled, bounded).

Show data table

Pearson correlation across 3 numeric columns (values clipped to 2 decimals).
	veteran_suicide_rate	civilian_suicide_rate	veteran_risk_ratio
veteran_suicide_rate	+1.00	+0.99	-0.86
civilian_suicide_rate	+0.99	+1.00	-0.91
veteran_risk_ratio	-0.86	-0.91	+1.00

state categorical label

This column contains US state names, with exactly 50 rows and 50 unique values — one row per state, perfectly uniform. Entropy ratio of 1.0 and a top_rate of 0.02 confirm complete uniformity with zero repetition, meaning this is effectively a lookup or reference table keyed by state. The 'long_tail' alert is a statistical artefact of the uniform distribution rather than a genuine concentration problem.

Treatment: Use as a join key or grouping label; no encoding needed unless joining to a larger fact table, in which case left-join on this field.

anthropic:default · confidence high

Out[12]:

saturn.columns["state"].stats

stat	value
n	50
nulls	0 (0.0%)
unique	50
top_value	Montana
top_rate	0.02
cardinality	50
entropy	5.644
entropy_ratio	1
alert: long_tail	50 singleton categories

Fig 7.

Top values for state.

Show data table

Top values for state (20 unique shown, of 50 total).
value	count	share
Montana	1	2.0%
Wyoming	1	2.0%
Alaska	1	2.0%
New Mexico	1	2.0%
Idaho	1	2.0%
Oklahoma	1	2.0%
Colorado	1	2.0%
South Dakota	1	2.0%
West Virginia	1	2.0%
Arkansas	1	2.0%
Nevada	1	2.0%
Arizona	1	2.0%
Oregon	1	2.0%
Utah	1	2.0%
Kentucky	1	2.0%
Tennessee	1	2.0%
Alabama	1	2.0%
North Dakota	1	2.0%
Missouri	1	2.0%
Kansas	1	2.0%

veteran_suicide_rate numeric numeric_target

This column represents veteran suicide rates, likely per 100,000 veterans, across 50 geographic or demographic units (probably U.S. states given n=50). All 50 values are unique with no nulls, indicating clean, granular measurement. The distribution is notably broad — ranging from 24.9 to 52.3 with a mean of 36.106 and std of 7.43 — meaning the highest-rate unit has more than double the lowest, a substantial disparity. The slight positive skew (0.42) and near-normal shape (kurtosis −0.77) with zero outliers suggest a relatively well-behaved continuous distribution without extreme anomalies.

Treatment: Use as-is in regression or ranking models; consider log-transform only if residuals show heteroscedasticity, as skew is mild at 0.42.

anthropic:default · confidence high

Out[15]:

saturn.columns["veteran_suicide_rate"].stats

stat	value
n	50
nulls	0 (0.0%)
unique	50
min	24.9
max	52.3
mean	36.11
median	35
std	7.426
q1	30.18
q3	41.33
iqr	11.15
skew	0.4169
kurtosis	-0.7696
n_outliers	0
outlier_rate	0
zero_rate	0

Fig 8.

Distribution of veteran_suicide_rate. Vertical dash marks the median.

Show data table

Histogram bins for veteran_suicide_rate (median: 35.0).
bin	count
24.9 – 28.81	10
28.81 – 32.73	9
32.73 – 36.64	9
36.64 – 40.56	8
40.56 – 44.47	6
44.47 – 48.39	4
48.39 – 52.3	4

civilian_suicide_rate numeric numeric_target

This column represents civilian suicide rates, likely per 100,000 population, across 50 distinct observations (possibly states, countries, or time periods). The distribution is remarkably well-behaved: near-zero skew (0.09), platykurtic shape (kurtosis −1.10), and no outliers detected, suggesting an unusually uniform spread across the full range of 7.7 to 28.9. The mean (17.618) and median (17.5) are nearly identical, and all 50 values are unique with no nulls or zeros.

Treatment: Use as-is in regression; near-normal distribution requires no transformation, but verify the unit (rate per 100k) before modelling.

anthropic:default · confidence high

Out[18]:

saturn.columns["civilian_suicide_rate"].stats

stat	value
n	50
nulls	0 (0.0%)
unique	50
min	7.7
max	28.9
mean	17.62
median	17.5
std	6.02
q1	12.6
q3	22.4
iqr	9.8
skew	0.08973
kurtosis	-1.103
n_outliers	0
outlier_rate	0
zero_rate	0

Fig 9.

Distribution of civilian_suicide_rate. Vertical dash marks the median.

Show data table

Histogram bins for civilian_suicide_rate (median: 17.5).
bin	count
7.7 – 10.73	8
10.73 – 13.76	8
13.76 – 16.79	7
16.79 – 19.81	8
19.81 – 22.84	7
22.84 – 25.87	7
25.87 – 28.9	5

veteran_risk_ratio numeric feature

This column appears to represent a risk multiplier or odds ratio specifically for a veteran population, with values bounded between 1.8 and 3.23 — all strictly above 1.0, suggesting it encodes elevated risk relative to some baseline. The mean (2.1714) sits above the median (2.0), and a skew of 1.09 indicates a moderate right tail, though no outliers were flagged. With only 41 unique values across 50 rows, there is light discretisation present, hinting at rounded or binned inputs. The relatively tight IQR (0.555) around a floor of 1.8 suggests the risk ratios cluster in a narrow band, which may warrant checking whether the 1.8 minimum is a data-entry floor or a genuine distributional boundary.

Treatment: Check whether 1.8 minimum is a hard floor/truncation artifact; apply log-transform to reduce right skew before regression modelling.

anthropic:default · confidence medium

Out[21]:

saturn.columns["veteran_risk_ratio"].stats

stat	value
n	50
nulls	0 (0.0%)
unique	41
min	1.8
max	3.23
mean	2.171
median	2
std	0.4021
q1	1.843
q3	2.397
iqr	0.555
skew	1.088
kurtosis	0.0735
n_outliers	0
outlier_rate	0
zero_rate	0

Fig 10.

Distribution of veteran_risk_ratio. Vertical dash marks the median.

Show data table

Histogram bins for veteran_risk_ratio (median: 2.0).
bin	count
1.8 – 2.004	25
2.004 – 2.209	7
2.209 – 2.413	6
2.413 – 2.617	4
2.617 – 2.821	3
2.821 – 3.026	2
3.026 – 3.23	3

How to cite

click to copy

BibTeX

@misc{saturn-data-trove-veteran-suicide-rates-2026,
  author       = {Steuber, Luke},
  title        = {Saturn reading: data trove veteran suicide rates},
  year         ={2026},
  howpublished = {\url{https://dr.eamer.dev/saturn/view/data-trove-veteran-suicide-rates}},
  note         = {Profiled with saturn-dissect v0.2.0, prompt saturn-insight-v2, model anthropic:default},
}

APA

Steuber, L. (2026). Saturn reading: data trove veteran suicide rates. Source: /home/coolhand/html/datavis/data_trove/demographic/veterans/military_firearm_suicide.csv. Profiled with saturn-dissect v0.2.0 (saturn-insight-v2, anthropic:default). Retrieved from https://dr.eamer.dev/saturn/view/data-trove-veteran-suicide-rates