Independent rankings built from public data — every figure traceable to the body that published it.
V-Dem Institute, University of Gothenburg · Political Violence by Non-State Actors
How often do groups outside the state use violence for political ends?
Of the 171 countries compared in 2025, Luxembourg had the lowest reading, -3.182, and Myanmar had the highest reading, 3.375. In the middle country the figure was -0.661. Among the 163 countries measured in every edition since 1990, the middle one moved from -0.495 to -0.663.
Part ofPeace & Security
Political violence by non-state actors · 2025
Key statistic · Peace & Security
Lowest of 171Luxembourg
-3.182012025 · position · derived
The three countries at the lowest end of 171 compared, the three at the other end, and the first on each continent.
1Photo: Michal Osmenda from Brussels, Belgium · CC BY-SA 2.0, greyscale
2Photo: Unknown author · CC BY-SA 3.0, greyscale
3Where a country sits says how it compares with the others measured, and nothing about why.
Seychelles-2.987
Uruguay-3.126
United Arab Emirates-3.134
Luxembourg-3.182
New Zealand-2.026
02Every measured country
Each country is shaded by its value, in five groups holding the same number of countries. Hover a country for its value, or select it to open its profile. Below the map, each continent’s middle country.
Darker is a lower value — the better end.
-1.320
the middle country, of 5 measured · 21 members
New Zealand-2.026
Papua New Guinea-0.328
Lowest and highest reading in Oceania.
-1.144
the middle country, of 40 measured · 49 members
Luxembourg-3.182
Serbia1.332
Lowest and highest reading in Europe.
-0.675
the middle country, of 48 measured · 52 members
United Arab Emirates-3.134
Myanmar3.375
Lowest and highest reading in Asia.
-0.381
the middle country, of 25 measured · 48 members
Uruguay-3.126
Venezuela2.652
Lowest and highest reading in Americas.
0.084
the middle country, of 53 measured · 55 members
Seychelles-2.987
Central African Republic2.976
Lowest and highest reading in Africa.
-2.114
the middle country, of 10 measured · 13 members
Iceland-3.108
Ireland-1.085
Lowest and highest reading in Northern Europe.
-1.890
the middle country, of 5 measured · 5 members
Turkmenistan-2.671
Tajikistan-0.060
Lowest and highest reading in Central Asia.
-1.507
the middle country, of 7 measured · 8 members
North Korea-2.780
China-0.506
Lowest and highest reading in Eastern Asia.
-1.457
the middle country, of 5 measured · 24 members
Trinidad and Tobago-1.799
Haiti2.317
Lowest and highest reading in Caribbean.
-1.122
the middle country, of 10 measured · 10 members
Czechia-2.761
Ukraine1.069
Lowest and highest reading in Eastern Europe.
-0.907
the middle country, of 13 measured · 17 members
Portugal-2.670
Serbia1.332
Lowest and highest reading in Southern Europe.
-0.891
the middle country, of 5 measured · 5 members
Namibia-2.750
Eswatini-0.588
Lowest and highest reading in Southern Africa.
-0.727
the middle country, of 7 measured · 9 members
Luxembourg-3.182
France0.343
Lowest and highest reading in Western Europe.
-0.658
the middle country, of 17 measured · 19 members
United Arab Emirates-3.134
Yemen3.086
Lowest and highest reading in Western Asia.
-0.454
the middle country, of 10 measured · 11 members
Singapore-2.219
Myanmar3.375
Lowest and highest reading in South-eastern Asia.
-0.204
the middle country, of 12 measured · 12 members
Uruguay-3.126
Venezuela2.652
Lowest and highest reading in South America.
-0.184
the middle country, of 16 measured · 16 members
Cabo Verde-2.606
Mali1.588
Lowest and highest reading in Western Africa.
-0.027
the middle country, of 6 measured · 8 members
Costa Rica-2.391
Mexico2.030
Lowest and highest reading in Central America.
0.011
the middle country, of 9 measured · 9 members
Gabon-2.010
Central African Republic2.976
Lowest and highest reading in Middle Africa.
0.078
the middle country, of 9 measured · 9 members
Sri Lanka-1.511
Bangladesh2.310
Lowest and highest reading in Southern Asia.
0.342
the middle country, of 18 measured · 19 members
Seychelles-2.987
Somalia2.715
Lowest and highest reading in Eastern Africa.
0.503
the middle country, of 5 measured · 6 members
Egypt-2.309
Libya2.296
Lowest and highest reading in Northern Africa.
2 of 2 measured — too few for a middle: Australia, New Zealand.
New Zealand-2.026
Australia-0.946
Lowest and highest reading in Australia and New Zealand.
3 of 5 measured — too few for a middle: Fiji, Papua New Guinea, Solomon Islands.
Fiji-1.815
Papua New Guinea-0.328
Lowest and highest reading in Melanesia.
2 of 4 measured — too few for a middle: Canada, United States.
Canada-0.939
United States0.752
Lowest and highest reading in Northern America.
03The whole field at once
171 countries, one dot each, on one scale — further right is a higher value. Half the countries are above -0.661; a quarter are above 0.482, and a quarter below -1.515. The bar in each row marks that continent’s middle country.
04All 176 ever measured
In order, highest value first — position 1 is the highest value, the way V-Dem Institute, University of Gothenburg’s own tables read. “=” marks a position shared with another country. V-Dem Institute, University of Gothenburg ranks nobody on this measure; the position is derived here by ordering the published values. ⚠ Higher is worse, so the derived position puts the most affected country first. It counts violence by actors outside the state; violence by the state itself is measured by the Physical Violence Index on this site, and crime committed for profit is excluded.
Showing 25 of 176 countries
Values for 2025. The change column compares 2025 with 2024 only; the sparkline covers the recent editions.
051990–2025 · 36 reference years
Each point is one year from 1990, all of them re-estimated together in V-Dem's v16 release. Rises tend to accompany insurgencies, contested elections and civil wars; the scale has no fixed ends, so read the movement rather than the number on its own.
| Year | Countries | Middle country | Middle half | Measured every time |
|---|---|---|---|---|
| 2025 | 171 | -0.661 | -1.515 – 0.482 | -0.663 |
| 2024 | 175 | -0.626 | -1.547 – 0.432 | -0.641 |
| 2023 | 176 | -0.617 | -1.543 – 0.356 | -0.650 |
| 2022 | 176 | -0.522 | -1.514 – 0.405 | -0.517 |
| 2021 | 176 | -0.530 | -1.561 – 0.423 | -0.539 |
| 2020 | 176 | -0.620 | -1.571 – 0.566 | -0.628 |
| 2019 | 176 | -0.620 | -1.682 – 0.704 | -0.628 |
| 2018 | 176 | -0.787 | -1.808 – 0.591 | -0.799 |
| 2017 | 176 | -0.651 | -1.623 – 0.591 | -0.730 |
| 2016 | 176 | -0.633 | -1.577 – 0.624 | -0.776 |
| 2015 | 176 | -0.757 | -1.577 – 0.591 | -0.776 |
| 2014 | 176 | -0.748 | -1.612 – 0.624 | -0.799 |
| 2013 | 176 | -0.755 | -1.684 – 0.500 | -0.799 |
| 2012 | 176 | -0.755 | -1.710 – 0.544 | -0.783 |
| 2011 | 176 | -0.737 | -1.737 – 0.523 | -0.742 |
| 2010 | 175 | -0.767 | -1.714 – 0.409 | -0.777 |
| 2009 | 175 | -0.767 | -1.730 – 0.259 | -0.798 |
| 2008 | 175 | -0.798 | -1.724 – 0.316 | -0.809 |
| 2007 | 175 | -0.801 | -1.743 – 0.285 | -0.809 |
| 2006 | 175 | -0.847 | -1.724 – 0.302 | -0.869 |
| 2005 | 175 | -0.743 | -1.724 – 0.388 | -0.749 |
| 2004 | 175 | -0.847 | -1.750 – 0.408 | -0.887 |
| 2003 | 175 | -0.818 | -1.713 – 0.471 | -0.826 |
| 2002 | 175 | -0.770 | -1.713 – 0.554 | -0.809 |
| 2001 | 175 | -0.760 | -1.713 – 0.559 | -0.791 |
| 2000 | 175 | -0.766 | -1.701 – 0.496 | -0.791 |
| 1999 | 175 | -0.732 | -1.523 – 0.637 | -0.757 |
| 1998 | 174 | -0.791 | -1.689 – 0.589 | -0.943 |
| 1997 | 173 | -0.757 | -1.691 – 0.593 | -0.770 |
| 1996 | 173 | -0.627 | -1.691 – 0.710 | -0.663 |
| 1995 | 173 | -0.627 | -1.663 – 0.761 | -0.645 |
| 1994 | 173 | -0.534 | -1.663 – 0.797 | -0.620 |
| 1993 | 173 | -0.534 | -1.513 – 0.842 | -0.620 |
| 1992 | 172 | -0.456 | -1.473 – 0.844 | -0.534 |
| 1991 | 170 | -0.577 | -1.451 – 0.938 | -0.620 |
| 1990 | 167 | -0.495 | -1.394 – 0.779 | -0.495 |
062024 → 2025
The biggest changes in value between the two newest editions — reference years recalculated in one report, not two separate publications. Places are shown second, because they also move when other countries do.
171 countries were measured in both 2024 and 2025. The field changed from 175 to 171 countries, so a place can move for reasons that have nothing to do with the country itself.
07Across every edition
The extremes of the statistic. A record reached by more than one country names none of them.
Highest reading on record
4.186
Original latent estimate (≈ −3.5 to +4); higher = more violence
Lowest reading on record
-3.427
Original latent estimate (≈ −3.5 to +4); higher = more violence
Biggest improvement between editions
−3.328
2.265 → -1.063
Biggest decline between editions
+3.570
-0.371 → 3.199
Most improved since first measured
−5.132
3.453 → -1.679
At the top most often
19 of 36
editions, alone at the first-listed end
08The measure
V-Dem's measure of political violence asks how often people and groups outside the state — insurgents, militias, party youth wings and the like — used physical force against people to pursue political aims. It is an expert-coded estimate on an open-ended scale centred near zero, and a higher figure means more violence.
What the number is: Original latent estimate (≈ −3.5 to +4); higher = more violence.
What it does not measure
Violence by the state's own agents is a separate measure, and crime committed for profit is excluded.
09Where it fits
Its part in the Country Index, the statistics whose values line up most closely with it across countries, and others on the same subject.
In the Country Index
A key statistic for Peace & Security
“Violence used for political ends by groups outside the state — militias, armed party wings and insurgents — the other side of the state violence that Physical Violence measures.”
Where political violence by non-state actors is higher, these tend to be higher too.
Where political violence by non-state actors is higher, these tend to be lower.
Rank correlation (ρ, from −1 to 1) across countries, on each statistic’s newest edition, out of 186 statistics sharing at least 80 countries with this one. Two statistics that move together do not cause each other.
10Keep reading
11Questions
Luxembourg, at -3.182, among the 171 countries compared in 2025.
Myanmar, at 3.375, among the 171 countries compared in 2025.
171 in 2025, and 176 at least once across all editions. The number measured varies from 167 to 176 between editions.
This site does not calculate a world average, which would depend on how countries are weighted. In the middle country of the 171 compared in 2025, the figure was -0.661.
No. V-Dem Institute, University of Gothenburg publishes the values but ranks nobody; the position on this page is derived by Nations Inquiry by ordering the published values, and equal values share a position.
This site holds 36 reference years, 1990–2025, all recalculated in V-Dem Institute, University of Gothenburg’s 2026 report on one method — not 36 separate publications.
From V-Dem Institute, University of Gothenburg, under CC BY-SA 4.0. From the V-Dem Institute's Country-Year Full+Others dataset (v16), University of Gothenburg — this indicator is not in the smaller Core file — published under CC BY-SA 4.0. The series here starts in 1990 so that every row belongs to a present-day country. Stored as the measurement model's original estimate, not V-Dem's 0–4 rescaling. Every edition is on this site’s Datasets pages as a full table, and downloads as a CSV file with every value unrounded.
12Method
Political Violence by Non-State Actors — V-Dem Institute, University of Gothenburg
From the V-Dem Institute's Country-Year Full+Others dataset (v16), University of Gothenburg — this indicator is not in the smaller Core file — published under CC BY-SA 4.0. The series here starts in 1990 so that every row belongs to a present-day country. Stored as the measurement model's original estimate, not V-Dem's 0–4 rescaling.
Newest edition
Report year 2025Data year 2025
Cite the source as: Coppedge, M. et al. V-Dem Country-Year Dataset v16. Varieties of Democracy (V-Dem) Project, University of Gothenburg.
Licence: CC BY-SA 4.0, checked 2026-09-26.
The file itself, column by column — where it came from, what each column holds, and every edition as a CSV — is on its dataset page.
The countries at the top of this statistic, from their profiles: Wikimedia Commons, shown in greyscale, each used under the licence named.
Nations Inquiry (2026). Political violence by non-state actors by country: Political Violence by Non-State Actors, data from V-Dem Institute, University of Gothenburg. https://nationsinquiry.com/statistics/political-violence